18650 battery pack design for a powerbank

Mmmkay, the above is true but I could also say “dividing the total capacity by the charging rate won't give you a correct charge time, but close”, and it too would be true. In assembled battery packs, where the resistance between the charging controller and the battery pack is very low, the duration of the CV charging phase is quite short.

So, in essence, I could say that dividing the total capacity by the charging rate in a battery pack with built in charging controller gives a very close charging time. If you disconnect exactly at the given time, the battery will be close to fully charged, or even quite close to 100% presuming a calm :-) charging rate. Charging a ≈20Ah battery at 2A is a rather calm, 0.1C-rate. :-))

I will just say I don't find the above tool very useful (it has some irrelevant, but disturbing floating point inaccuracy), but if you do, use it. The involved arithmetics are fairly easy, you can do them with a regular calculator or even mentally.

Technically speaking the charging board does not detect the battery reachs 4.2V, as it is not monitoring battery voltage right at the battery terminals; or at least I wouldn't say it this way. Once the charging board reachs close to 4.2V at its output, it starts tapering down the current to the battery or, more precisely said, the current tapers down as a consequence of the output voltage getting frozen or stuck at the maximum allowed battery voltage:

If you look at the current curve, once it starts going down at minute ≈153 the voltage curve is at 4.16V. Many chargers do this early tapering scheme, which results in a slightly longer CV phase.

The actual charging cut-off is (usually) set slightly below 1/10 of the charging rate; look at the curve and you will see the cut-off, close to the 270 minute mark, at ≈60mA.

If you look at the capacity of the cell getting charged, ≈3000mAh or a hair above, and divide it by the 1A charging rate we obtain 3ish hours. Look at the 180 minute mark in the graph, and observe the effective capacity of the cell at that point. I see no less than 2.8Ah, which is 93.3̅%.

Advice: filling a battery to the brim takes a lot more time and is also more stressful for the battery. Fill the tank less and enjoy both faster charging and better battery life.

The above graph is from one of HKJ's TP5000 board reviews, check them here and there.

Interesting CC/CV module, looks well heatsinked.

The CC/CV module won't stop charging, i.e. it won't stop delivering current because it will keep the output at the set voltage indefinitely (current flow tends to 0 over time). This is no problem if you plan on stopping the charging by hand at some given time, but it would be rather bad for the battery if you were to leave it charging for a lot more time than needed (undue high voltage stress). If setting output voltage at rather low values, 4V - 3.95V or less I'd say, it may not matter at all.

Nah, it didn't really work. It limited the output current, but started tapering very early.

I say it didn't really work because there was no cut-off, the chip never entered the CV phase as it seems hard-coded at 4.2V (close to 4.2V, in my experience).
I am sure it would still limit current if fed somewhat below 4V, but according to what I see in the indicator leds it needs a minimum of 4.25V to work (onboard blue led only turns on at that input voltage and up).

Hello Barkuti,

Thanks for your reply, I learned lots of things :D

Still in this thirst for learning, I have several news questions :

1) About the TP5000/TP5100 module, a few day ago, we talked about the heat they can released during the charging process. To get 2A you advised me to combine two module in parallel set-up at 1A each, Today I ask myself a question, Is it really reliable to do this? Can they cause some weird bouncing effects due to the two ICs fighting each other?

2) About the CC/CV module and the TP5000/TP5100, does it exist an external component like an relay that can measure the voltage at the out charging terminal and when per example 4.2V is detected the component cut-off the charging module supply?

3) Another question about how to optimising TP5000/TP5100 module heat cooling. I thought of adding a Heat sink on top of the chip, however, I have a doubt about the efficiency of this one, knowing that the chip (tp4056) is 4mm * 4mm, the heat dissipation may be limited. So I was thinking of adding a larger Heat sink under the board with an adhesive thermal pad for heat conductivity, what do you think?

4) Currently, I'm designing a little 18650 charger with an tp4056 module, I would like to track the voltage and amperage during the charging process. To do this i add this little antmeter, but i'am not sure if i can wiring like this? Here is my wiring diagram, What do you think ?

Ps: I decided to put the voltmeter powering wire at the input of the tp4056 to not to disturb the charging process.

Diyfr,

1) While I didn't try with TP5000/TP5100 modules, these are from the same company as TP4056 units. And TP4056 modules happily work in parallel.

2) Mmm, not as far as I know. If you stop the CC/CV module from receiving power once its output reachs 4.2V, the battery won't be fully filled as it would prevent the CV phase from happening (which could be good lifespan wise).

3) Installing a heatsink over the back copper plane is fine and effective. At 1A the modules will work without problems, though.

4) It's fine.