5v Power Bank Module

1 view

Skip to first unread message

Laszlo Perry

unread,

Aug 5, 2024, 4:16:25 AM8/5/24

to rairaalomhe

Ihave just taken delivery of a batch of USB C 2A powerbank charger + boost circuits from a seller on AliExpress. China churns out millions of these circuits every year, so you'd think they'd be good at it by now.

So long story short, I connect one of these up to a pack of 18650 cells that I have rescued from an old laptop battery. Connect the USB C port to a suitable power supply (through my trusty Keweisi USB power monitor) and everything appears to be good. It starts flashing an LED at me to say it's charging and So now for even more digging. Looking around at videos from BigClive (you should check his channel out, especially if you like drunk Glasweigians) it looks like the chip (which they have handilly laser etched away the number on the top) is the good old IP5306 which has a recommended circuit of:

It's the lack of R2 that is most concerning. Without that the chip has no way of knowing what is going on with the battery. So time to do another little modification - the addition of a resistor. I don't have a 0.5 ohm one to hand, but I do have a 1 ohm one, which should do just as well. I can always put two in parallel if I really need to.

These modules, which are from a seller called "TZT Official Store", are obviously some reject or prototype that went wrong, and should really not be being sold by them. But if you have been duped into buying any from them all is not lost - these two small modifications should get them out of the rubbish bin and back into something useful.

It seems all I can get from these "2A" modules is 1.5A max. Pretty poor really. Even after adding a second 1 ohm resistor to take it to the recommended 0.5 ohm, and adding the missing capacitor across the battery (C2 in the schematic). And then the inductor on the board gets rather hot. Maybe it's time to upgrade that inductor to something a little bigger.

I'm not a technician, but I can understand something about electronic. So I searched many topics in many foruns, and electronic sites about batteries and chargers. The result is few answers and many doubts.

My idea would be buy one module and use some 18650 batteries. Nice. Or not?

Well, I asked to sellers how many batteries can I use in each module. One told me 4, other told me there isn't a limit. What do you think? Many batteries will taken a long time to charge, but it's ok. But, will the module show the correct state of charging? Can the module deal with different charging state of each battery?

I extended my doubts to the problem of connecting batteries in parallel. When the module is off, the problem of one good battery charging a bad battery and it creates a great current could occour? Will I need to connect resistors among batteries? Would it occour even with protected batteries? (in principle, I want to use protected ones)

So I'm afraid to create a bomb instead of a power bank.

- -single-usb-pcb-board-plastic-enclosure-for-power

The first problem I found in these products is my protected 18650 batteries don't fit the space for batteries. Protected batteries are bigger. But, my fear about parallel batteries wouldn't occour even with these power bank too?

I don't know if the idea of making a good and safer power bank is practicable for me.

So, is there a good module in market that can deal with some batteries, even in parallel? Do you recommend this kind of DIY project? (I've saw some guys here making thein own power banks and it seems to be all right with them).

If I give up the idea, maybe the best option I found is TOMO V8-4. Although some guys had experienced not good results with it, Maybe it can be a nice choice ( $ 13.49 now in gearbest). And, complementing, you know Soshine E3 (almost the same as TOMO V8-4). In discussions and reviews here we could see it need some improvements. Do you know the new version E3S ( ) ?? The LCD is different, and maybe there are software and hardware improvements, but I cannot find any review of it.

Thanks guys! Sorry for being so long.

Rafael, you can set up as many cells as you'd like in parallel; in such a case, unless you make something very wrong, they'll operate as a single battery with a capacity of the sum total of all of them, and an equivalent internal resistance equal to the inverse of the sum of the inverses of each battery internal/lead resistances (sort of the sum of all power delivery throughputs).

Additionally, can these kind of modules deal with batteries with different state of charging? Is there a problem in case I use batteries of different mAh, or brands, or state of charging? Do I need to balance the batteries before turn on the first time?

Regarding differences in capacity, brands or even chemistries, I really doubt this could be an issue because of the low power delivery requirements for powerbanks. Just don't use worn out, Ultrafakeor other known PoS cells (plenty of cheap brand name units nowadays).

Powerbanks can draw 1 amp or even more from each cell in a 4P configuration under heavy loads, it is ALWAYS advised to use the same capacity, model and most importantly chemistry, because each can have different discharging curves and voltage recovery when not under load.

While I am not going to deny the potential concerns regarding all of that battery mismatching stuff, and speaking here about chemistries of course, I bet you my kingdom two or more brand name batteries of different chemistries but at least comfortable enough power delivery margins would happily operate in parallel without issues. Properly connected, any voltage differential caused by faster discharging through the lower IR chemistries should be rapidly and continuously rebalanced because of the then inherent dV between the cells. I am not saying, in any way, that I support this but, because of the lack of actual testing data regarding these matters, this stuff could work well. Maybe not for n00bs/rookies, of course.

The first FastTech link in the opening post shows a really cool module I tried. Solid performance and the LCD reports battery voltage plus current output. Uses two independent current sensing resistor arrays, one for each port. I bridged them in parallel to slightly increase the efficiency/diminish voltage drop without issues. Handy. A bit expensive, though.

1'2Kg of weight? Well, that means, at the very most, just 2118650 cells could be inside of it (3S7P arrangement). Now, if these cells were to be top of the line VTC5As or VTC6s, we would be speaking of 10'8V nominal, 18'2/21Ah peak capacity and about 210A of continuous discharge, with higher peak value yet with subsequent larger output voltage drop.

I have a voltage regulator to 3.3V that feeds a circuit with a PIC and a LoRa Module, the circuit transmits for 10 seconds and remains sleep for 5min, has a total current consumption of 16mAh. If the circuit is powered from the USB port of the PC it works continuously, but if I connect it to a Power Bank it only works the first 5 minutes. What can I do so that the Power Bank does not shut down?

ProgrammerID:

Apparently it usually works based on peak current, rather than average; a common technique I've seen is pulsing the dummy load every few minutes, for like a few milliseconds at a time. This is an issue that lots of people have wrestled with and there's a lot written on it. A bit of time with google will find info on what other people do about this.

Also 16mAh is not an amount of current. Current is measured in mA. Capacity of a battery is measured in mAh (milliamp hours, ie, a "100mAh" battery could supply a 10mA load for 10 hours, or a 100mA load for 1 hour).

Well, if you could put a 100 ohm resistor across the output of the power bank to create a 50mA load (Ohms law). There is then not much point in activating sleep mode in your Arduino, if that was seen as an energy saving measure.

You may find that there are more efficient ways of preventing the power bank doing a shutdown.

6v6gt:

You may find that there are more efficient ways of preventing the power bank doing a shutdown.

What you usually do in a case like this is switch that dummy load with a MOSFET, and pulse it on briefly every time you wake up (and use the WDT to wake every 8 seconds for this purpose) - in the vast majority of things with autoshutoff, a brief pulse of the load periodically is enough to keep them awake.

DrAzzy:

Exactly. Post #6 and #9 here Powering the arduino nano with a Power bank - Project Guidance - Arduino Forum go in this direction, but it requires some experiment with the specific power bank to minimise the wasted energy.