some questions for SunKing, and anyone else on lithium battery technology

[gbynum]

I r edecuated as an ingineer, but emphasis was in electroinics and RF ... so you can use techy terms ok. I'll probably misunderstand them. My degree was in 1972.

Google is my friend .. but not a very good friend. We don't communicate very well.

I'm fascinated by the graphs showing energy stored as a function of charge voltage. If I read them correctly, there is very little additional energy stored as cell charge voltage increases from 3.(5 or 6 or 7 depending on chemistry) to some "maximum, perhaps 4.1 or 4.2. But I don't see various chemistry addressed. Since it is what I have to "play" with, let's address "standard" 18650 cells as in a flashlight. Via data acquisition stuff I have, the charger I have shuts off at about 4.13V. I'm gathering that the charger is pushing a little too hard? Would I really get ALMOST equal energy stored at 3.9V and get better cell life?. This is somewhat rhetorical; I have never "fully" discharged a cell in a month, and swap a charged for a used cell each month.

Battery University helps a little; Panasonic's website should, but I cannot fine the correct place.

So ... if I use perhaps 20% of a cell (or battery) capacity in a month and recharge ... would a somewhat lower (say 3.9 vs 4.1) shutoff make any difference?

Where can I find authoritative data?

THANKS! to you experts.

[sensij]

18650 is the size of the battery, but doesn't tell you its chemistry. Assuming standard Li-ion, it looks like you can significantly extend cycle life by lowering the voltage you charge to, if you can tolerate the capacity hit. lithium.JPG


More detailed information here:

http://batteryuniversity.com/learn/a...ased_batteries

Note that these voltages do not apply to LiFePO chemistry, so make sure you know what you have before you mess around too much.

[Sunking]

As Sensij says 18650 is just the size of the case it is in, not what is inside. There is no reason to ever charge any lithium battery to 100% SOC. For power tools no big deal because the scale i sso small, the manufacture makes a lot more money off you selling you a new battery every year vs every 3 to 4 years.

Things change when the scale goes up. EV manufactures never allow the customer to fully charge the lithium battery. It is the only way they can offer the warranty they do. Additionally the risk of fire is greatly reduced. Those Panasonic Cells Tesla uses only have 500 cycles in them if charged to 100%. At 80 to 90% have 2000 cycles.

The idea of fully charging batteries is a throw back to lead acid batteries. To maximise Lead Acid battery life you must fully charge them immediately after use and keep them fully charged. Lead acid batteries do not work well in a Partial State of Charge. Lithium batteries work best in a PSOC, exact opposite of lead acid.

To answer your question of how to determine SOC depends on which lithium type you are talking about. There are also a few ways to charge lithium. So you have to be more specific. Lithium cobalt cells charge/discharge curve is steeper than a flatter LFP and thus easier to determine. LFP works best using a Float type charger of roughly 3.4 vpc will get you up around 90%.

So be specific. LTO is around 2.5 vpc, LFP 3.3 vpc, LCO 3.6, and LMC 3.7.

[gbynum]

Thanks to both of you. I'd browsed batteryuniversity but missed that page. Sunking, a small part of my question is how to determine what chemistry I have. I recognize I'm totally inexperienced; the 3 cells I have are protected 18650 between 2.3Ah and (supposedly) 3.0Ah, and are used in single-cell flashlights. I instrumented a charge cycle by a commercial (Nitecore) inexpensive charger, 2 position capable of 1A split between 2 if 2 are being charged; I didn't monitor current, but the charger charges in pulses, LIKELY monitoring voltage between charge pulses to determine shut down. Since it went to (I don't have the data file at my fingertips) about 4.12 IIRC, and they are called "lithium ion", I'm guessing LCO or LMC.

They are charging at a maximum of 0.4C, so temperature rise is slight. They shutoff before 4.2 so single charge damage appears unlikely. But with 3 cells and 2 lights ... and never exhausting one in a month, I'm averaging under 9 cycles a year, so 90 (10 years) will probably outlive me and they'll die of something else. I've data files available if anyone is curious but I ran my DAQ at a slow rate to have a small file with up to a 6 hour charge and don't see millisecond resolution of the charge pulse.

THANKS again.

George

[sensij]

Any labeling at all on the cells you have?

[karrak]

If the cells are around 2.3Ah-3Ah the are most likely to be LCO cells. LFP cells have about a 1/3 lower energy density. The maximum charge voltage for LFP batteries is 3.65V/cell.

As a general rule keeping the SOC of lithium ion batteries centered on 50% for intermittent use will give you the best life. This is difficult to do using voltage with LFP batteries because the SOC vs voltage graph is so flat.

It is an old article but there is some good information here tesla.com/blog/bit-about-batteries, sorry about not posting a clickable link, for reasons that are beyond me I am not allowed to post links.

Simon

Off grid 24V system, 6x190W Solar Panels, 32x90ah Winston LiFeYPO4 batteries installed April 2013
BMS - Homemade Battery logger github.com/simat/BatteryMonitor
Latronics 4kW Inverter, homemade MPPT controller

[Sunking]

If the cells are around 2.3Ah-3Ah the are most likely to be LCO cells. LFP cells have about a 1/3 lower energy density.

Complete hogwash Karrak. Anyone can search and see you are full of it. Wanna try again. LFP has a higher Power Density than all LCO with the exception of hybrid exception of a hybrid LiPo used in RC models with a Gel Electrolyte. At best you can make a case for some hybrids having almost 200% more, but 300% is laughing stalk material.

[Sunking]

It is an old article but there is some good information here tesla.com/blog/bit-about-batteries, sorry about not posting a clickable link, for reasons that are beyond me I am not allowed to post links.

Real simple Karak you are in the same club as Dan. It is because you are full of crap and the mods will not allow you to post links. They are giving you a hint which you should take.

[sensij]

Complete hogwash Karrak. Anyone can search and see you are full of it. Wanna try again. LFP has a higher Power Density than all LCO with the exception of hybrid exception of a hybrid LiPo used in RC models with a Gel Electrolyte. At best you can make a case for some hybrids having almost 200% more, but 300% is laughing stalk material.

In this chart, LCO has an energy density of 165. 1/3 of 165 = 55. 1/3 lower than 165 = 165 - 55 = 110. LFP is just under 110. I think Karrak may have been looking at the same chart as what you posted, describing it in different words than you would use.

[gbynum]

Any labeling at all on the cells you have?

Yes, but I doubt it helping much. As I said, these are sold for flashlights; 2 of my 3 are NITECORE NL183, 2300mAh 3.7V 8.5Wh with overcharge and discharge protection. I use a Nitecore charger. While not my need, they are "High discharge performance" "Li-ion Rechargeable Battery". I'd guess my maximum discharge rate is under 1C ... light is advertised at about 2 hours at maximum briteness, but they don't specify what cell(s) are used for that ... it can use a pair of CR123A as well. From Sunking's list, I'd guess these are LMC at the 3.7V.

I don't smoke either, but have observed that these are used in those vape things.

What I get from these posts is that if I observe (assumed constant current pulsing) charge voltage and end charging at 4.0 vs 4.15, I'll perhaps lose 10% in capacity and perhaps double cell life from charge/discharge effects. 3.9V might be a better stopping voltage.

Oh, and Sensij, I have no intention of making my own charger, rather the quest for knowledge. I did enough explosive manufacturing in my youth.

[gbynum]

So be specific. LTO is around 2.5 vpc, LFP 3.3 vpc, LCO 3.6, and LMC 3.7.

Is LMC what BatteryUniversity calls NMC?

[sensij]

According to a thread over in Candlepower forums, that Nitecore battery is built around a DLG N18650 2200 mAh cell. Based on the information available from DLG, that probably means NCM chemistry, under the NMC heading on batteryuniversity.

http://www.dlgbattery.cn/en/upload/2016-08/12/DLGyuanzhuxing.pdf

[gbynum]

Thanks! Staying off the solar panel subject (do the mods care ... this is a battery subforum), my next step is to understand protection circuits. I do visit candlepower but find more technical info here, more emotional there ...

When I hook up my DAS to the assembled cell terminal, the protection circuit is in the charging path ... how much drop is attributed to it? Time to use Google.

[Sunking]

I am not going to bother looking up the actual part number to tell you what type it is. I will tell you how to figure it out yourself. The key is Nominal Voltage

3.2 volts = LiFeP04 aka LFP or Lithium Phosphate. Charge 3.6
3.6 volts = LiCoO2 aks LCO or Lithium Cobalt Oxide. Charge 4.2
3.7 -3.8 volts = LiM2O4 aka LMO Lithium Manganese Oxide. Charge 4.2
3.6 - 3.7 volts = LiNiMnCoO aka NMC Lithium Nickel Manganese Cobalt Oxide. Charge 4.2 , What a mouth full of alphabet soup.
3.6 volts = LiNiCoALO2 aka NCA Lithium Nickel Cobalt Aluminum Oxide. Charge 4.2 volts
2.4 volts = Li4Ti5O12 aka LTO Lithium Titanate. Charge 2.85 volts. These are the monsters of Lithium batteries that can last up to 10,000 cycles. They have low Energy Density (wh/Kg) as indicated by the voltage, but the highest Power Density W/Kg and W/L. These cells can be charged and discharged very fast, and the safest of all battery chemistries. Almost no channce of thermal runaway.

Here is your takeaway. Does not make a lot of difference what type it is from a manufacture charging POV as there are only 3 voltages to take note of to 100% SOC: They all Charge at C/2 and Terminate C/33. Note some can be charged faster than C/2 up to C5. But C/2 and lower is safe for all types.

2.85 for LTO
3.6 volts for LFP
4.2 volts for everything else. LCO,LMO,NMC, and NCA

Hope that helps you get started. Those batteries are either LFP, or NMC. So all you need to know is what voltage of either 3.6 or 4.2 volts.

[gbynum]

Thanks Sunking. So my original question; existing charger cuts off at 4.15 or so ... given that I don't need full energy stored, would I be better off to stop charging at 3.9 or 4.0 ... assuming I can determine how to do so?