Looking for maintenance and care advice for Lithium ion battery pack

[Gauss163]

The driving question in my mind is at what discharge rate should the 50% state of charge be measured at...? [...]

Only in very extreme applications will that matter much, e.g. the jump-starter packs that you mentioned. I manage my jump-starter packs very differently from other Li-ion batteries.

[...] My round LiIon cells are around 7 - 8 years old. I have optimally stored them at 4.0 volts. They are still working very well...

Why do you consider 4.0V to be optimal? 4.0V is about 80-85% SOC for common chemistries, which will likely cause more degradation than storing them in the recommended ranges of 35-50% SOC.

[SilverFox]

Hello Gauss163,

Inquiring minds want to know...

At what voltage do you store your jump packs at?

And,

Would drone and RC helicopter use be classified as extreme use?

I define "optimal" by combining cell degradation and intended use. My stored cells are ready to go in an emergency without needing to charge first, and 4.0 volts is easier on the cell than storing at 4.2 volts.

An argument could be made that in the best interest of the cells I may be better off investing in more cells and storing them at a lower state of charge. While I have pondered that my current regime seems to be working well so I most likely will continue.

I test my cells annually at 1C and compare the results to when the cells were new. I recycle my cells when their capacity drops below 80% of their initial capacity.

Tom

[Gauss163]

At what voltage do you store your jump packs at?

In the winter I store them at SOCs that will yield a few jump starts at lowest ambient temperatures.
In the summer I store them between 40-50% SOC.

Would drone and RC helicopter use be classified as extreme use?

They differ in that jump-starters need to be on standby for use at unpredictable times, whereas drone use times are usually far more predictable (often scheduled), so one has the opportunity to exploit the known use times to develop much more optimal storage charging schemes, e.g. storing at optimal voltages, then topping up as close to use-time as proves convenient.

[SilverFox]

Hello Gauss163,

If I have a drone that has 20 minutes of run time and fully charge it and then run it for 10 minutes, I should be at 50% state of charge. Even if the resting voltage of the pack is showing 3.85 volts per cell.

Correct?

Tom

[Gauss163]

If I have a drone that has 20 minutes of run time and fully charge it and then run it for 10 minutes, I should be at 50% state of charge. Even if the resting voltage of the pack is showing 3.85 volts per cell. Correct?

The recommended storage voltage ranges are designed to minimize the degradation caused during the portion of calendar life spent in storage. These ranges are designed to not be so high that that the high voltage accelerates internal degradation processes, nor so low that the pack could self-discharge to low SOC levels that are also unhealthy.

Suppose we have a high-current device such that the pack capacity at the high-current rate is only half of the nominal (C/5) capacity. In your example, if you discharge it at the high-current rate to 50% of the high-rate capacity, then you are only discharging 25% of nominal capacity. So when the pack reaches resting voltage it will be at 75% SOC (typically 3.90-3.97V). That is far from the optimal recommended ranges. Jump-start packs are an even more extreme example of this difference between nominal and very-high-rate capacity.

[novarider]

Gauss, I'm still waiting on a link to a "top-tier" lipo manufacturer that recommends storing their batteries as low as you say they should be.

[Tachead]

Hello Gauss163,

If I have a drone that has 20 minutes of run time and fully charge it and then run it for 10 minutes, I should be at 50% state of charge. Even if the resting voltage of the pack is showing 3.85 volts per cell.

Correct?

Tom

I recommend you store your drone's lipo packs at the recommended voltage provided by the manufacturer. It is usually written on the pack or available at the manufactures website. And yes, your method works well. Fully discharging your packs until low voltage cutoff may shorten their life and most hobbyists generally only like to discharge to about 3.7-3.75V. Most lipo pack manufactures recommend a storage voltage between 3.7 and 3.9V from my experience so the usual storage mode of 3.85V on most hobby chargers works well and is convenient if you miss the mark with your flight. I store my packs at the manufactures storage recommendations and have gotten years of service and many cycles out of them. So have the many other RC hobbyists I know and fly with.

Gauss is no expert(unless you consider Googlefoo a skill that one can be considered an expert of), he just regurgitates information and charts gathered from Wiki, Battery University, and HKJ's site and likes to think he is more knowledgeable then engineers and experts from multi-million dollar battery manufacturers. The information he has provided in this thread is mainly for different types of cells from different manufactures. He has even stated that the information and recommendations from most lipo manufactures is erroneous but, offers no proof and seems to think his word holds more weight then the hundreds of engineers and experts that some of these companies employ. I wouldn't trust anything he says personally but, YMMV.

[Gauss163]

Gauss is no expert (unless you consider Googlefoo a skill that one can be considered an expert of), he just regurgitates information and charts gathered from Wiki, Battery University [...]

You know absolutely nothing about my professional background, yet you continue to make completely unfounded personal attacks such as the above. Furthermore, what you wrote makes it quite clear that you aren't even bothering to pay close attention to what I write, since anyone who does knows quite well that I don't consider Battery University to be a reliable source. That speaks volumes about how completely unreliable your claims are. If you wish to dispute some claim then the proper way is not to make (unfounded) personal attacks on the author but, rather, to provide links to reputable (scientific) sources that support your views.

You cannot possibly gain deep knowledge about battery electrochemistry from rough generic guidelines in hobby charger manuals, battery pack labels, etc. Continuing to stubbornly interpret such as the bible on battery electrochemistry is not the path to enlightenment. Nor is behaving in puerile ways that will alienate knowledgeable users that attempt to share their professional expertise in hobby forums.

[Tachead]

You know absolutely nothing about my professional background, yet you continue to make completely unfounded personal attacks such as the above. Furthermore, what you wrote makes it quite clear that you aren't even bothering to pay close attention to what I write, since anyone who does knows quite well that I don't consider Battery University to be a reliable source. That speaks volumes about how completely unreliable your claims are. If you wish to dispute some claim then the proper way is not to make (unfounded) personal attacks on the author but, rather, to provide links to reputable (scientific) sources that support your views.

You cannot possibly gain deep knowledge about battery electrochemistry from rough generic guidelines in hobby charger manuals, battery pack labels, etc. Continuing to stubbornly interpret such as the bible on battery electrochemistry is not the path to enlightenment. Nor is behaving in puerile ways that will alienate knowledgeable users that attempt to share their professional expertise in hobby forums.

I may not but, the fact that all the information and charts you post come from Wiki, HKJ's site, etc. instead of scientific publications, text books, and research studies speaks volumes imo. I try not to pay close attention to what you write because me along with several others(as you can see from this thread alone) have realized you are just another know it all armchair cowboy with too much time on his hands that thinks he is an expert because he can use google. I don't have to support my views Gauss. I don't claim to be an expert like you and my view is simply that following the manufacturers recommendations instead of some know it all on the internet is a better and safer course of action. The only one who really knows how their cells should be cared for is the manufacturer because they are the ones that made the cells and are the only one who knows the actual specifications, chemistry, and engineering of their particular cells.

I said nothing about "gaining deep knowledge of battery electrochemistry" nor that "hobby charger manuals and battery pack labels are the bible on electrochemistry". I simple said that a manufactures labels should be followed because they made the cells. There are thousands of different cell designs and variations of a particular cell chemistry(that is why cells even of the same general chemistry don't all perform the same) and the manufacturer is the only one who knows these specs. So, why you would suggest going against their recommendations is beyond me. It is careless and egotistical to make such recommendations and assume your knowledge is superior to the hundreds of engineers and technicians that R&D, produce, and test their cells.

[nbp]

Oooook. This thread is becoming a debate between just two members. OBVIOUSLY, you are not going to convince each other. You've each said your part, let's leave it there and let some other folks carry on the discussion. Thank you in advance for your cooperation.

Remember when I posted this?

Stop trolling each other. Seriously.

[Gauss163]

For readers who are actually interested in the science behind these matters, below are some graphs from a recent study that show how degradation typically depends on storage SOC. This is for NCA chemistry, whose degradation properties are less severe then other common chemistries. Just as I described above, you can see how the internal resistance and capacity fade increases at higher storage SOC levels.

Excerpted from: Comparison of Plug-In Hybrid Electric Vehicle Battery Life Across Geographies and Drive Cycles, by Kandler Smith et al., National Renewable Energy Laboratory, 2012.

[SilverFox]

Hello Gauss163,

While I agree with what you are illustrating with you supposition, it is flawed.

You are illustrating what happens when you use a high capacity cell in a high current application. High capacity cells tend to have higher internal resistance and that cause them to heat up during high discharge rates. The heat degrades the cell rapidly.

Let's change the supposition...

Suppose we are using high current cells in a high current application. Take a look at these cells.

It looks like 3 minutes at 20 amps will result in the same 50% state of charge that 30 minutes at 2 amps will. After 3 minutes at 20 amps if the resting voltage ends up at 3.85 volts, we have cells at 50% state of charge with a voltage of 3.85 volts.

It is interesting to observed that there is a recent push to store high current LiPo battery packs at 40% instead of 50%. 40% state of charge has a resting voltage of 3.7 volts.

Tom

[Overclocker]

Hello Gauss163,

If I have a drone that has 20 minutes of run time and fully charge it and then run it for 10 minutes, I should be at 50% state of charge. Even if the resting voltage of the pack is showing 3.85 volts per cell.

Correct?

Tom

And yes, your method works well.

20min runtime implies high current draw. depending on the drone's LVC (low voltage cutoff) you might not get all the juice out from the battery because high currents induce a large voltage drop, so LVC might be reached prematurely.

a better way would be to do a FULL discharge using a hobby charger at 1c or lower. note the capacity. then do another discharge that only drains half of the capacity. that would be your 50% soc

depending on the chemistry e.g. NCA going from 50% to 60% storage SOC already entails a significant hit in capacity loss, at least according to this study.

[SilverFox]

Hello Overclocker,

I agree that it is important to know how your particular application works. With the cells that I referred to low voltage cut off was not a concern.

Looking at the charts you posted it looks like the difference between 50% and 60% state of charge is a little under a 1% drop in capacity. This brings up the question if a 1% drop in capacity over 10 months is significant to my use...?

Keep in mind that my storage is usually for a 5 days or less with flying almost every weekend and my storage temperature is always below 25 C.

Tom

[Overclocker]

well i've already stated my position on the other page. for small packs it really doesn't matter. when you get to the larger vehicle packs that you may intend to use for years then small differences do add up

almost OT, i just finished wiring this up:

[Gauss163]

While I agree with what you are illustrating with you supposition, it is flawed.

Please be more precise: what "supposition" do you disagree with?

You are illustrating what happens when you use a high capacity cell in a high current application.

Neither true nor relevant. It seems that you are referring to the example I gave in this post, in reply to your question about SOC after running a drone for half of its usable capacity. Since it seems that the point of that example was not clear, I will elaborate below.

The point was to give an example where the usable capacity is much lower than nominal capacity (at C/5 rate). Here the usable capacity refers to the capacity we obtain under actual use of the device, where the (average) current may be so much higher that then nominal (C/5 rate) that the pack yields much lower capacity than the rated nominal (C/5) capacity (e.g. this is typically true for the jump-start packs that you mentioned, and many other extremely high-current devices).

In such extremely high-current apps, when the pack is at 50% of usable capacity, there still may be much higher nominal (chemical) capacity remaining (but to access all of it we'd need to discharge at a much lower rate). In such cases it is only the nominal (chemical) capacity that is closely correlated with the resting (open-circuit, no-load) voltage of the cells during storage. It is this resting voltage that governs many of the degradation processes during storage. Thus SOC levels for storage guidelines always refer to percentage of nominal (better chemical) capacity, not the possibly much lower usable capacity.

Suppose we are using high current cells in a high current application. Take a look at these cells. It looks like 3 minutes at 20 amps will result in the same 50% state of charge that 30 minutes at 2 amps will. After 3 minutes at 20 amps if the resting voltage ends up at 3.85 volts, we have cells at 50% state of charge with a voltage of 3.85 volts.

But that's not an example of an extremely high-current app because the tests don't go up to high-enough current that the usable capacity starts diverging from the nominal capacity. If you take any cells to high-enough current they will eventually reach some point where the usable capacity starts dropping off dramatically (mainly because the IR drop will push the flat (low-slope) part of the discharge curve below the termination voltage, so much of the capacity will be delivered below termination voltage level).

It is interesting to observed that there is a recent push to store high current LiPo battery packs at 40% instead of 50%. 40% state of charge has a resting voltage of 3.7 volts.

Yes, as I said, generally 40-50% SOC is considered a good storage level (but the exact voltages that correspond to those levels depend on the chemistry). Due to competition the RC/hobby manufactures are now being forced to provide more optimal numbers - even though doing so significantly cuts into their profit margins. This is exactly analogous to what occurred with laptop batteries, e.g. most all now have "battery saver" software, and analogous features have recently started to appear in the RC world, e.g. recent DJI packs have a user-configurable "auto-discharge" feature that will drain the pack to 40-50% SOC after a couple days idle (the number of days idle is user-configurable). Before this feature appeared power users jury-rigged their own devices to discharge DJI packs to what they deemed to be healthy storage levels, e.g. look up the Phantom Angel.

[Gauss163]

[...] depending on the chemistry e.g. NCA going from 50% to 60% storage SOC already entails a significant hit in capacity loss, at least according to this study.

Thanks for providing further scientific data supporting my statements about the relationship between storage degradation and SOC. What is the source of those graphs? (they look familiar but I can't place them off the top of my head).

[SilverFox]

Hello Gauss163,

My illustration was to show that some cells handle 2C discharge rates without serious degradation of capacity.

I showed cells that can handle 10C discharge rates without problems.

If the drone battery pack is constructed with cells like this and only subjected to 2C discharge rates, then running the drone for 10 minutes of the 20 minute run time will in fact discharge the cells to 50% state of charge.

I don't believe at 2C discharge qualifies as extreme.

To be perfectly clear...

I design an application that involves a 2C discharge rate and supply it with cells that can handle 10C without degradation in capacity. Running the application for half the run time will reduce the state of charge to 50%.

Back to the drone. With an expected run time of 20 minutes I can reach 50% state of charge by running the drone for 10 minutes. This assumes that the cells in the battery pack are high current cells that do not suffer degradation in capacity at 2C.

If after 10 minutes of run time the battery pack shows a resting voltage of 3.85 volts per cell, then 50% state of charge is 3.85 volts per cell.

Keep in mind that the original question involved storing for a week with flying on the weekends. My comment is to fully charge the battery pack and run it for half the run time and then keep it at a temperature lower than 25 C. Do you have any studies that show that storing for 5 days under these conditions will severely degrade the battery pack?

Once again to be perfectly clear...

Battery pack is made up of cells that are capable of 10C discharge without a significant loss in capacity.

Application runs at 2C.

Storage is for 5 days.

Storage temperature is less than 25C.

Tom

[Gauss163]

@Norm, Yes, if you are discharging at a rate that yields capacity close to nominal, then 50% of capacity at that rate will be close to 50% of capacity at the nominal rate. However, you need to keep in mind that if the load is not constant current (e.g. constant power), then 50% of runtime need not be 50% of capacity.

[Overclocker]

Thanks for providing further scientific data supporting my statements about the relationship between storage degradation and SOC. What is the source of those graphs? (they look familiar but I can't place them off the top of my head).

http://jes.ecsdl.org/content/163/9/A1872.full

[Gauss163]

For convenience, below is a summary of the primary degradation / aging mechanisms of the anode and cathode in Li-ion cells. Notice in particular how high SOC plays a role in the majority of the degradation processes.

[Gauss163]

Excerpted from: Predictive Models of Li-ion Battery Lifetime, by Kandler Smith, NREL, 2014.

[Gauss163]

To expand on my remarks in post #5, here is some further data on the degradation caused by float-charging and storing at high SOC and high temperatures. Below is a graph from this 2014 study of calendar life degradation (no cycling) for Sanyo UR18650E cells stored at various SOCs at 50°C. As the study shows, you get only 107 days lifetime storing at 100% SOC, but 2.7x that (288 days) storing at 95% vs. 100% SOC, and 14x that (4 years) storing at 0% SOC.

During calendar aging tests cells were stored at different SOC and temperatures. Fig. 6 shows capacity fade and resistance increase for cells stored at 50°C at different SOC. As expected, cells stored at lower SOC exhibit a longer lifetime. Especially cells stored at 100% SOC show a much faster degradation compared to other SOC. These cells only reach a lifetime of 107 days at 50°C until capacity reaches 80% of the initial value (linear interpolation). At this time the resistance increased by a factor of 1.3. For comparison, cells stored at 95% SOC reach 288 days until end of life (linear interpolation), whereas a cell stored at 0% SOC has an extrapolated lifetime of about 4 years (linear extrapolation of the last three measured data points)