tp4056 Series Input power, parallel output (to battery)

[rollmeister]

Hi,

I am waiting for my tp4056's to arrive and purchased several however after watching a video on Youtube, was wondering if I can apply them in my final application because of a potential common grounding issue.

Basically the module supports 4-8v input to charge one lithium battery to 4.2v however my power source is 20v and hoped I could connect the inputs of three tp4056's
https://www.aliexpress.com/item/Mini-USB-5V-1A-Lithium-Battery-Charging-Board-Charger-Module-In-4-8V-TP4056/32522176440.html
in series (dividing the 20v into 6.66v input for each tp4056), then connect the output to battery in parallel thereby working as a step down dc-dc converter one in go and increase max charge rate to 3amps.

Output of the board perhaps is not isolated, and cannot be done?

Using a multimeter, can I test for this with negative probe to input negative of the first module connected in series to the 20v source, then probe the positive output of each tp4056 board and if finding a potential, it would confirm shared ground?

Regards,
Rollmeister (electronics newbie)

[sleemanj]

No that will not work, a complex IC full of silicon gates and logic designed for some random task (in this case charging lithium cells) does not a voltage divider make.  Use a buck converter to take your 20v down to something suitable and feed your tp4056's in parallel (each charges it's own  battery no other connection on output side between them!!)

[Inverted18650]

If you parallel the 4056 you can increase the output current to 3A. The layout is in the datasheet of the TI model I think...I’d try to remember to check back here when I am at my PC and if no one else has linked it in, I’ll get it for you.

[soubitos]

The basic TP4056 module without protection has 4 terminals.
+VIN - VIN +VOUT - VOUT
You can parallel as many as you like to increase charging current.
Their behavior is quite random though depending on component tolerances and what happens (based on trial and error) is that if you have say 3 in parallel and you are charging a 3P pack of 18650's, provided you have enough input power, it will start charging with max current all three "red" charging indicator leds will light and as soon as current drops, one of the three modules will switch off and show charged... as charging progresses, the second module will also show charged and then when the pack is fully charged, the last one will also show charged.
If you use the following configuration in a stand alone charger that's that, you charged your batteries and all is ok... If you try to use the combination though to also power a load, two things will happen... 1. depending on the current you draw, the TP4056's will need to also pass enough current to power your load and will get hotter than normal for longer time and at the same time, the charging time will be much longer and 2. the batteries will never complete charging and at least one of the modules will be in charging mode constantly.

As for the input, you do need a back converter or voltage regulator to bring the voltage within specs for the TP4056... TP4056 is using a drop resistor in its input (i use 0.43Ohm, seen 0 ohms !! though in many of the chinese modules) and you should not exceed 5-5,5 volt if you want to use the modules you got at full charging current.


If your modules have output protection, you can stack them one on top of another with pin headers but their actual operation is even more hectic as there are more components involved and more "tollerances"... it will work... but not quite as you might expect.

What you describe, ie 3 modules in series with parallel output will NOT work


To go around and between these problems and issues, i have designed two modules TP4056FLEXADV and TP5100FEXADV... The TP4056 module is basically a combination of a back converter, charging and protection circuit and a boost converter in the output... the USB plug is reinforced (you will find out VERY soon that the usb terminal on the chinese pcbs is only one pull away from flying of the board, hopefully without shorting your power source), you can use a screw terminal and dc-jack to power the module allowing for a wide range of power sources and you can use screw terminals to connect to your load directly from the battery (with protection of course) or via the booster circuit.... There is a jumper to select charging current, no need to go through replacing a tiny smd resistor to do that... Input voltage can be either 5V-USB or 8-45V (i wouldnt go more than 24 to keep it safe unless charging current is less than 500mA) and output voltage is V-bat up to 24V 200-800mA or directly from battery.

TP5100 is a more expensive, and more complex module, allowing for 4.5-18V (20max not recommended) input without extra components and max charging current is 2A. Charging current here is fixed (I figured, if you need less than 1A, go with TP4056, no need to "waste" a more expensive chip to do the job of TP4056) and there is thermal protection for the batteries as it is wise to keep it safe and protect the average 18650 cell from overheating as they dont like 1C charge currents for prolonged time.

I expect pcbs and parts for the latest version of TP4056FLEXADV (V2.2F) which is improved and offers extra features too.... Unfortunately, i have sold out, otherwise i would be happy to send you a board.

[rollmeister]

Thank you for the replies.

This all boils down to my lack of knowledge and experience. I am a web developer, not an electronic enthusiast and I went into this assuming too much.

20v to about 5v dc-dc buck down should be acceptable for me, efficiency from AC mains input is not important and these are cheap to buy.

I still think using the same tp4056 in parallel should still be ok, and the lithium batteries all have bms and are connected in parallel also.

[soubitos]

Going from 20V down to 5V and charging with max current via the TP4056 is not a "cool"thing to do LOL.....
In my first batch of TP4056 modules i used an LM7805CV .... powered from 8V and charging with 1A it reached 100C
If you cannot introduce a separate 5V power supply (a phone charger is always 5m away from any living being on the planet LOL) make sure to at least use a switching power supply or be prepared to sink a LOT of heat...


If you parallel TP4056 modules in order to increase charging current, think about it twice too.....
I did it and it did not work that well... charging took much longer than expected when charging 2-3-5P packs, much better when charging single cells though... It would help to give us an idea about how many and what of cells you will parallel.. also if they are "new"or "recycled"batteries with varied capacity....

Then you mention BMS in single series many parallel cells pack... ?? What do you need a Balancing circuit with a single series pack anyway? Perhaps you mean the cells are protected? This does might complicate things even further i think....

[Inverted18650]

Here is a link for the 3A units ebay.

https://www.ebay.com/itm/1PCS-TP4056-4-2V-3A-High-Current-Lithium-Battery-Charging-Board-Charger-Module-M/152348735563?epid=834798867&hash=item2378b1384b:g:ZhgAAOSwnHZYSTIZ

I bought a roll of 100 of the generic TP4056's for $5.99 a while back and made up a bunch of these myself. This thread has got me thinking about doing a little video series about the topic. It is certainly possible and works very well and is very stable. Though, like the others have mentioned, try to get VIN closer to the drive voltage of the chip.

[soubitos]

OP's first problem is his noted 20v power supply... the higher the charging current (provided the 20v supply can handle it) the higher the heat on any type of voltage regulator to power TP4056... even if you bring this 20v supply down to a safe (for the TP4056) 7V range, then the TP4056 will also heat up significantly (try touching a single TP4056 when at full 1A load and you will feel my concerns  ).


Then, he is probably getting TP4056 modules not chips....
Frankly, if I wanted to charge a 1S and 1-2P pack, i would use a single TP4056 anytime... even at shared charging current, the batteries will charge in a matter of a few hours anyway.... If not powered from 5V USB then i use a switching regulator to minimize heat problems.... TP4056 will not be happy when at full 1A charging current with anything above 5V and that, with a 0.2-0.5ohm drop resistor as seen in the datasheet.

Now, if i wanted to charge 1 or more cells in parallel with higher current up to 2A and with a higher input voltage, TP5100 is the way to go... it is a much more advanced build esp for someone without experience soldering SMD esp. QFN packages (been there, hurt a lot)....


Until now, i have not found any other charging chip which is cheap and can do the job over 2A.
To charge a beefy 1S10P pack for a safety night light inside a store room i used one of my TP4056 modules which also powers the rest of the circuit and took some 60hrs to terminate charging (also due to recycled cells).....



If i had to charge a 1SxxP pack for a more demanding app, i would use a CC-CV power supply on the cheap and a protection board for the pack.. I dont think its wise to manipulate chips made for x purpose that much.... like the 3A TP4056 modules which as far as i picked from others on youtube etc are prone to overheating among other issues. In my case, i stacked 4 TP4056 modules with protection one on top of the other with all IO in parallel, it worked very well as a charger but had other issues as a power source for the application after it.

[Inverted18650]

The TP5100 is a good chip,  and I like your EasyEda project too. I run the 4056’s at about 750mA each and use 4 of them, “borrowed” the design directly from the eBay unit. I might try to string 20 of them in a row and see what happens...lol got them so damn cheap who cares if they burn up.

[soubitos]

Well... if you use them with half their rated amps and make sure their input voltage is under 5V they will be cool..
What i am trying to put together (but fail miserably doing) is to use a TP4056 with some sort of external power handler (mosfet?) so that the chip is controlling voltage and charge cycle but the power is handled by the mosfet/s

[biskero]

Hello,

I wanted to ask for some help on the tp4056 modules https://www.aliexpress.com/item/5PCS-Micro-USB-5V-1A-18650-Lithium-Battery-Charger-Board-With-Protection-Module/32647794396.html?spm=a2g0s.9042311.0.0.LWh0eB

I have the following scenario for a ebike project. The battery pack is a 10s10p 18650, total 36volts 25Ah.
So I need to charge the battery and wanted to see if the TP4056 can be used.
I was thinking to use 2 or 3 in parallel to charge each 10P battery cell.
From my research this is ok as long as the 10P battery cells are not connected to each other which is not my case as I have 10S of this 10P cells.
In addition I want to be able to charge the battery while the ebike is using it, i.e. solar/dynamo charging while riding.

Is this correct? Or Am I miss something?
Any suggestion on how to design such charging solution? I know I can use one of those ebike specific balance charging module but I found out that the cells are not really balanced and wanted to see if there is a better solution.

What do you think?
Thx


PS: found this might be the solution

[soubitos]

So you need to charge 10P cells of a 10S pack with TP4056... first of all, you need to maintain isolated ground between the cells charging "4056" .... that alone is a no no and no again.
Then you need to understand that to effectively recharge even partially a 10P portion of the pack you need at least a few amps, with 5A per S block you would need at least 5hours to recharge it more in real life. a lot more... to achieve a 5A charging current from a stack of TP4056 modules you need 5 x 10S you need to interconnect 50 modules at least... good luck with that...  note, i do NOT say it is not possible, i say it is NOT PRACTICAL .... add the 10 ground isolated power sources to this and i am 100% sure the cost already exceeds that of a dedicated charger for the job.

[biskero]

Hello,

that's what I am investigating as I already have a balance charger board for the ebike but from my tests are not the best as components wise.
Isolating the TP4056 charging negative from battery negative as shown on the video with a b0505s module is possible but as you mention how practical would be!?
The cost is cheaper than a balance charger module for 18650 so that's not the problem but I agree with you is a messy solution.

Need to do some more investigation on different solution.
Thx for your feedback and if you have other suggestion pls share it, I am open to ideas.
Thx

PS found a new module 3A https://www.aliexpress.com/item/3-7V-4-2V-3A-Li-ion-Lithium-Battery-Charger-Protection-Board-For-18650-TP4056-DD05CVSA/32824653351.html?spm=2114.search0204.3.37.9c1d2cd2p8jGnU&ws_ab_test=searchweb0_0,searchweb201602_5_10152_10065_10151_5722813_10344_10068_10342_5722613_10343_5722913_10340_10341_10698_10697_10696_10084_10083_5722713_10618_10304_10307_10302_5711215_10059_10184_10534_308_100031_10103_441_10624_10623_10622_5711315_5723013_10621_10620_5722513,searchweb201603_19,ppcSwitch_5_ppcChannel&algo_expid=e457c8bc-2f6f-4795-9827-7cf046015171-5&algo_pvid=e457c8bc-2f6f-4795-9827-7cf046015171&priceBeautifyAB=0

[soubitos]

This is a protection board normally incorporated to tp4056 modules

[soubitos]

And a VERY expensive one too.. with 1.24$ you can make them yourself and keep the change