When I say Lithium Ion, I am referring to LiCoO₂ LiMn₄O₂ and LiNiMnCoO₂ cylinder batteries. When storing lithium ion for some time, never keep it topped off. Better 3.7ish. Many of the data sheets I've studied lists that the batteries are rated to handle anywhere from 300 - 2000 cycles. Usually, the longer they last, the lower the energy density of the cell. This is typically a full 4.2 v charge and a 2.50 - 3.00 volt discharge. Don't drain a Lithium Polymer cell lower than 3.0 volt though. Lithium Ion cells love the partial charge and discharge cycles it documented that this can expand their life cycles by as much as 4 times their rated. Lithium ion differ from NiCad and NiMh cells, draining and fully charging them is just a waste of energy. Some Chinese manufacturers list this in their instructions but they usually sell rebranded cells that do not perform at its advertised specs. LiFePO4 does its own things and holds less capacity but is much safer and uses different voltages too.
I've heard some say that it is a good practice to fully charge and fully discharge a power bank to calibrate their capacity display gauge. It makes sense if it checks and calculates the current voltage and both total input and output current.
Yes, I agree. I was only talking about "topping off" if the battery is in regular use. Lithium-ion batteries should be stored at about a 40% state of charge. If you look back through this thread I mention and refer to most of the same points you make here.
It's also true that li-ion batteries should definitely not be drained lower than about 2.5-3 volts, as you say. As for discharging li-ion power banks, (or protected cells) you can drain them down "all the way" because the circuitry will cut off the discharge at the correct low voltage.
It's interesting about discharging and charging power banks for the first few cycles. When I purchased my first power bank about five and a half years ago many people on Amazon were complaining that the bank wouldn't hold anywhere near the rated charge claimed for it. Many returned them. I noticed this too the first time I used it. However, this particular power bank ( a Chinese 20,000 mAh generic) came with a small card in the package that advised the buyer to fully drain and then fully charge the bank for the first three cycles. I was skeptical when I first heard about this because it seemed to make no sense. Nevertheless, I went ahead and tried this and was astonished at the change in performance. Suddenly, the bank came to life and started performing really well and it would hold a charge under load for extremely long periods of time; basically performing according to its specs. Maybe it was just recalibrating but compared with other devices I own this felt different. And I wonder why if it was just a calibration they would recommend doing it three times in a row. Anyway, the bank lasted about five years in nearly daily use running two dash cams simultaneously and in fact is still functional, although at this point it probably has about 1/3 its original capacity. It sits on the shelf at about 40% charge.
So, I don't really have an explanation for why cycling the power bank for the first three cycles made such a difference but I recall seeing a reference somewhere to the notion that lithium batteries like to get cycled after long periods in storage. I think it was on
Battery University. I'll see of I can find it.
I also agree with you about not fully charging lithium batteries. My understanding is that they should be charged to about 85%. Truth be told I often just let the banks I use charge fully because in my experience with quite a few of them now they last many years with excellent performance and are really quite inexpensive to replace these days. The most recent TQKA 20,000 mAh LiFePO4 I purchased was only $16 for example, so if I happen not to get the ultimate number of potential cycles out of it because I let it charge to 100% I'm OK with that.
BTW, today I was testing the TQKA and running two Mobius cameras it dropped to 10% from fully charged in about an hours time. As I was out of state and about 50 miles from home I just let it run all afternoon and into the evening and then when I got home it was still showing 10% and plenty of juice left in it. So, it seems the percent-of-charge display has a problem but the battery cells and charge circuitry seems fine. At the same time I was also running two other dash cams on my other TQKA LiFePO4 Pentagon. That one started out the day at 100% charge and ended up at 83% when I arrived home.