DIYBMS for Lithium Iron Phosphate battery cells (LiFePo4) 280aH

48v is a nominal figure in the same way a 12v car system is (usually around 14v) so 16 cells are normally used and ‘48v’ equipment is designed to operate on that.

With cheap cells, make sure you top balance and if possible do a capacity test so that you can match the parallel cell strings to be as even as possible.

imho - generally the cheap ones are ‘Grade B’ or rejected ‘Grade A’ where their capacities or internal impedances are not within spec for the electric car/truck/everything else markets.

Make sure you return anything with physical damage or swelling

TBH as you are not cycling them and using them as stand-by there should not really be any issues

1 Like

hello,
here you can see pics of my setup, still in builtup phase but running fine, I’m impressed from all the work done for this project.
Communication is running fine w/ 2P16S on 32 boards. I’m using 9k6 bouds and have cycle times from 1700ms, not many lost packets. You can see two blue light running thru the boards :slightly_smiling_face:
every thing was setup step by step until I wanted to test balancing.
While balancing is active on any board, the communication stops there.
So seem that the balancing fet causes so much disturbance that communication brakes down. Once the cell finished balancing the communication comes back.

Did anyone experience the same?
Did someone experiment in extending the switching time of balancing fet e.g. turn it on for some seconds and than off?


I’m running 16s2p but with two controllers, ie each 16s is standalone.
This gives me the ability to take one battery offline without losing the whole system.
I tested the 9k6 SW but went back to the standard speed. I can’t remember if I tested it to 100% SOC?
The recent controller SW is a considerable improvement in reliability for me.

Looking at your photos where are you getting power for the controller?
I found it best to power it from a 60V buck converter via the 5V input rather than the ESP32 microUSB.

At the moment I power the ESP32 via the microUSB with a USB Power adapter. On long therm I have same idea to be prepared for offgrid.

Better to use a supply with common ground reference, ie derived from the battery.
Try connecting that supply to the 5V input instead.

I am just wondering if someone working with EVE_1.9 project upgrade for V450 (with new controller) . Today I received 8x 230Ah from CATL and I am suprised that EVE_1.9F also fit perfectly.

1 Like

For me senseless to upgrade to 4.5 with 1624 as long as the controller board is not capable to program the 1624.

How high is the balance current in the 1.9f modules?

The original boards can do 0.8A.
The 1.9f can handle 1.2A.

But be aware. This is a passive system.
It’s not balance. It’s burning. You produce heat instead of transfer.

yes I know that. I was only curious. A friend said his Battrium modules burn 2A and I couldn’t tell him how much the 123mm LFP burns. Now I know it :slight_smile:
I also connected active 4A neey BMS in parallel to the DIY BMS.

I work similar. Have the DIY and in addition a JK with 2A balancing.
Advantage DIY:: Webserver. Easy to use in the home net.
Advantage JK: energy transfer.

Both together get my cheap LFK cells usable balanced.

How do you wire 2 packs batteries in series parallel?

Hello
I have created own PCB from module v4.5 fitting EVE LF280K cells by changing dump resistors to 4x4x 1R1 and Q1 to a bigger one (3,14A@3,45V).
Now i have read sourcecode of AtTiny1624 and have seen, that Voltage level for balancing is hard coded to 4.00V - on LiFePo4 i have to change it to a new level like 3,40V OR 3,45V or can it be adapted by controller? Are there any good values for PID-Controller on those Big Cells?
Furthermore the mAh counter is fixed - so i have to adapt it too or not?