Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
TP4056 type C and 118650 pack
#1
I am experimenting with 1S4P 18650 powered by TP4056 Type C and be able power devices using the TP board output points and at the same time recharge the batteries using the input points.

The TP's will manage each cell individually for over charge, discharge, heat and many more functions and avoid the use expensive and sometimes unreliable BMS.

Because my PCB design have the batteries are sitting in 4x holders there is no bars, solder or fuses all the work is manage by the TP board that has been design specially to manage the 18650 batteries.

My idea is to build a flexible powerwall easy to manage by making a 1S20P board (Or bigger) to obtain 58 Amp (or more) and 4.1 Volts per board then I can combine as many as I want to make a battery pack.

The PCB boards will be charged by solar panels at 5V and possibly from 2 to 10A each PCB battery board regulated by a small buck power controller to ensure the input voltage is maintain at 5V

Also I am investigating the option to have a group of small solar cells of 5V 2A each to power each PCB independently with their own mini USB power controller (Only investigating the possibility)

All this will be set on my PCB designs ...... this is a sample 1S4P and is working correctly using my phone charger at 5V 1A to maintain the batteries at optimus charge at all times. 

Will love you gurus to give me some feedback good or bad .... will help me a lot 

Check out my test PCB and set up test bed 

Thanks a lot


Cesar


Reply
#2
I take it that you want to be able to swap cells a lot?
Otherwise you could just go for a large single charge controller (eg one big TP4056 type device) charging all the cells in parallel?
Korishan likes this post
Running off solar, DIY & electronics fan :-)
Reply
#3
I see a few problems with this setup as a powerwall pack design:

* TP4056 chargers are very inefficient for long term use
* They can get very hot
* They don't all charge at exactly the same voltage/rate
* You'd need a LOT of them
* The cell holders are "very" bad for any kind of real load due to bad contacts. They have high IR and cannot make good solid contact with the cell

Suggestions:
* If you want to stick with individual cell chargers, go with TP5100's.
- They can take in up to 12VDC making them more efficient on the input wire side
- They are a buck converter, not linear regulator, so a lot less heat output
- And are more accurate on their output voltage/current
- TP5100's can charge at 2A. So you could go with 2P or even 4P with 1 board. This would lower the overall cost/parts needed

(note: even though they are rated for 2s, there may be some modifications needed to achieve this successfully)


* Cell holder change to a different type, ones that have brass/copper contacts or beryllium springs. You may have to order the springs separate and replace the stock ones for beryllium ones
- There are "SMD/SMT" versions of the cell holders, and those usually come with brass contacts and better springs to hold the cells
Sed6 likes this post
Proceed with caution. Knowledge is Power! Literally! Cool 
Knowledge is Power; Absolute Knowledge is Absolutely Shocking!
Certified 18650 Cell Reclamation Technician

Please come join in general chit-chat and randomness at Discord Chat (channels: general, 3d-printing, linux&coding, 18650, humor, ...)
(this chat is not directly affiliated with SecondLifeStorage; VALID email req'd)
Reply
#4
(06-28-2020, 10:26 AM)Redpacket Wrote: I take it that you want to be able to swap cells a lot?
Otherwise you could just go for a large single charge controller (eg one big TP4056 type device) charging all the cells in parallel?
????? how will you make a big TP4056 device ????? - the reason of my design is for future maintenance when I have to replace some cells .... my way is easier just take them out and insert a new one and because the cells on the PCB are all in parallel they should balance correctly ..... am I correct ??? on my assumptions ???? pls. let me know

(06-28-2020, 12:49 PM)Korishan Wrote: I see a few problems with this setup as a powerwall pack design:

* TP4056 chargers are very inefficient for long term use
* They can get very hot
* They don't all charge at exactly the same voltage/rate
* You'd need a LOT of them
* The cell holders are "very" bad for any kind of real load due to bad contacts. They have high IR and cannot make good solid contact with the cell

Suggestions:
* If you want to stick with individual cell chargers, go with TP5100's.
- They can take in up to 12VDC making them more efficient on the input wire side
- They are a buck converter, not linear regulator, so a lot less heat output
- And are more accurate on their output voltage/current
- TP5100's can charge at 2A. So you could go with 2P or even 4P with 1 board. This would lower the overall cost/parts needed

(note: even though they are rated for 2s, there may be some modifications needed to achieve this successfully)


* Cell holder change to a different type, ones that have brass/copper contacts or beryllium springs. You may have to order the springs separate and replace the stock ones for beryllium ones
- There are "SMD/SMT" versions of the cell holders, and those usually come with brass contacts and better springs to hold the cells

Thanks a lot for the info and yes during my test run (1S4P) i notice the TP gets hot for small time then they cool down .... but yes they get hot and the specifications indicate that they should resist up to 80C ???? maybe not ... Thank you I will look at the TP5100 .... maybe is a better solution .... what mods will I have to make ????
Thank you very much
Reply
#5
Make sure you post your replies "outside" of the [quote] blocks, plz Wink makes it easier to read

Redpacket means for one large TP device is a single charger, not a bunch of smaller ones. There is no such thing as "balancing parallel cells with multiple chargers". It can't happen. When cells/batteries are in parallel, they are balanced, just by the very nature of being connected in parallel.

Heat is an enemy, at any level, with battery construction. (we're excluding extreme cold temps here). So if the TP unit, or any unit for that matter, is producing heat, it is wasting energy. The reason the TP unit gets hot then cools off after awhile, is because the cell is getting close to charged. There's less voltage potential between input/output on the TP unit so it doesn't have to waste so much energy.
A linear regulator takes a voltage input of say 5V and outputs at a particular voltage, lets say 3V in the case of a discharged cell. 2V's worth of energy is dissipated as heat (the actual energy wasted is in watts, not volts. But the amount of watts is specific to the amount of current flowing). If current is 500mA, then 2 * .5 = 1W. If you have 100 of these per parallel block, that's 100W. Then you have 7s, that's 700W. If you're building a 48V system, that's 14s or 1400W of heat generation. That's a LOT of wasted energy. 1400W is the equivalent of a space heater

TP units are great for single cell application where you are doing batches of cells to charge up to do capacity testing, or for single device applications just as a rechargeable flashlight, toy, etc. They are "horrible" for use where lots of cells are involved.

Ohhh, also, something you didn't think about. For every parallel pack in the series (ie, for 7s, 12s, 14s) you'll need a "different" power supply. You can't connect all the TP units together to a common ground. They'll all get dead short and pop every one of them.
Proceed with caution. Knowledge is Power! Literally! Cool 
Knowledge is Power; Absolute Knowledge is Absolutely Shocking!
Certified 18650 Cell Reclamation Technician

Please come join in general chit-chat and randomness at Discord Chat (channels: general, 3d-printing, linux&coding, 18650, humor, ...)
(this chat is not directly affiliated with SecondLifeStorage; VALID email req'd)
Reply
#6
Korishan ... It is great help UR giving me thank you .... I have been looking at the TP5100 and looks great for what I want to do I will take in consideration your advice and continue researching and see if my thinkings take me o a Ok solution (With your provide advice - Thanks) .... maybe back to BMS ???? and soldering Ummmmm hard work but maybe that is the best way .... I will buy a few TP5100 and see what happens ....Thank you
Sorry I am a new user so I will make mistakes using the site he he he he
Reply
#7
Just looked at those PCBs again. That's a lot of wasted space on those. I hope they were only prototypes and you got a few. Cuz if you planned on making them all that size, that would of been expensive to make enough.


As far as the cell holder. I'm not knocking using them, tho they are not as reliable as soldering/spot-welding. You just need to have good enough connection on the tabs. Nice stiff springs that provide very little resistance. i think someone else here actually took a coil spring and added them to the tabs, between the case and the tab, to help give it more pressure.
Also, "always" take cells out of those holders from the NEG end "first". Otherwise you take the chance of making Pos/Neg contact on the Pos end when pulling them out. I've done it and many others have as well
Ibiza likes this post
Proceed with caution. Knowledge is Power! Literally! Cool 
Knowledge is Power; Absolute Knowledge is Absolutely Shocking!
Certified 18650 Cell Reclamation Technician

Please come join in general chit-chat and randomness at Discord Chat (channels: general, 3d-printing, linux&coding, 18650, humor, ...)
(this chat is not directly affiliated with SecondLifeStorage; VALID email req'd)
Reply
#8
Thanks again mate yes the board is big an only prototype for me to play and maybe add something else if needed he new design is just a bit longer than the holder plus the TP and the side is 2 mm bigger than the holder.
Yes I take negative first I got my finger burn one time for taking from the positive and make a full Pos/Neg contact on my finger and burn he he he he the heat was instant and little time to pull out .... you are correct mate Neg first ....

thanks again
Korishan likes this post
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)