FBMS - FERCSA's bms

FERCSA

Member
Joined
Jan 16, 2017
Messages
87
Hi!

I started thinking about a very simple cell balancing system. Of course passive, but smart. Basically every pack gonna get a balancer module then daisy chained together. So only 3wires needed because of the i2c line and gnd. The power come from the pack, before you ask my guess is the consumption in idle should be around a few mA. Then all data goes to the master module for data collection, monitoring, display on a web interfaceetc.. Option for configuring every balancer module from the master.

The specs is the following:
-it can be used in stand alone mode aka top balancing the cell on default.
-bottom balancing is a option if master is present.
-configurable over voltage and low voltage threshold
-isolated data(i2c) connections!
-max. 2A or 4A balance current, both are possible because the selected mosfet can handle it.
-more than one temperature sensor for safety
-cheap-o, my guess is around 6-7EUR per module, but definitely under 10EUR.

What do you think? Any suggestion? Any otherfunctiondo you like to see?
OrI should scrape the whole thing then reverse engineer the longmon's protocol? :D
 
Definitely, do your own work would be quicker and for 10EUR for longmon is it worth the effort to rip off someone else's work?
 
hbpowerwall said:
Definitely, do your own work would be quicker and for 10EUR for longmon is it worth the effort to rip off someone else's work?

Ohh Pete. It was a joke (smile doh). Looks like it's disappeared on the ocean somewhere between europe and australia. I have a weird humor I know and it has already been proven.
 
FERCSA said:
hbpowerwall said:
Definitely, do your own work would be quicker and for 10EUR for longmon is it worth the effort to rip off someone else's work?

Ohh Pete. It was a joke (smile doh). Looks like it's disappeared on the ocean somewhere between europe and australia. I have a weird humor I know and it has already been proven.

Welll, it is a big ocean, i think the :D must have turned in to a :dodgy: on its journey.....
 
FERCSA said:
Hi!

I started thinking about a very simple cell balancing system. Of course passive, but smart. Basically every pack gonna get a balancer module then daisy chained together. So only 3wires needed because of the i2c line and gnd. The power come from the pack, before you ask my guess is the consumption in idle should be around a few mA. Then all data goes to the master module for data collection, monitoring, display on a web interfaceetc.. Option for configuring every balancer module from the master.

The specs is the following:
-it can be used in stand alone mode aka top balancing the cell on default.
-bottom balancing is a option if master is present.
-configurable over voltage and low voltage threshold
-isolated data(i2c) connections!
-max. 2A or 4A balance current, both are possible because the selected mosfet can handle it.
-more than one temperature sensor for safety
-cheap-o, my guess is around 6-7EUR per module, but definitely under 10EUR.

What do you think? Any suggestion? Any otherfunctiondo you like to see?
OrI should scrape the whole thing then reverse engineer the longmon's protocol? :D

I was going to do almost exactly that with a few Arduinos. Later on down the track I can add extra functions, one that springs to mind is temperature monitoring.
 
FERCSA said:
Hi!

I started thinking about a very simple cell balancing system. Of course passive, but smart. Basically every pack gonna get a balancer module then daisy chained together. So only 3wires needed because of the i2c line and gnd. The power come from the pack, before you ask my guess is the consumption in idle should be around a few mA. Then all data goes to the master module for data collection, monitoring, display on a web interfaceetc.. Option for configuring every balancer module from the master.

The specs is the following:
-it can be used in stand alone mode aka top balancing the cell on default.
-bottom balancing is a option if master is present.
-configurable over voltage and low voltage threshold
-isolated data(i2c) connections!
-max. 2A or 4A balance current, both are possible because the selected mosfet can handle it.
-more than one temperature sensor for safety
-cheap-o, my guess is around 6-7EUR per module, but definitely under 10EUR.

What do you think? Any suggestion? Any otherfunctiondo you like to see?
OrI should scrape the whole thing then reverse engineer the longmon's protocol? :D


Are you thinking of using I2C to communicate between each cell BMS and the master BMS? If so then you must put in opto-couplers on the RX and TX lines as they will otherwise short out the arduinos via the common ground on the serial bus.

I like the idea of having it very modular and having a arduino at each cell could be very interesting down the line as it's very easy to expand on functionality later, you could even make them 'universal' by making them accept other voltage ranges than just 2.8V - 4.2V so they could be used for other battery chemistries too.

Hey you could even make each arduino manage a number of cells where the number of cells would be a function of available ADC pins and I/O pins for controlling mosfets and so on.

You would need to come up with a resistor network for lowering the cell voltages to something the ADC pins on the arduino can handle and obviously when you scale the cell voltages down you will loose out on the precision of the ADC's.
 
mormor1971 said:
Are you thinking of using I2C to communicate between each cell BMS and the master BMS? If so then you must put in opto-couplers on the RX and TX lines as they will otherwise short out the arduinos via the common ground on the serial bus.

Exactly! Yes I know that as I stated on my list number 4 "-isolated data(i2c) connections!"

mormor1971 said:
I like the idea of having it very modular and having a arduino at each cell could be very interesting down the line as it's very easy to expand on functionality later, you could even make them 'universal' by making them accept other voltage ranges than just 2.8V - 4.2V so they could be used for other battery chemistries too.

Actually this was my main goal, modularity and versatility. Because I plane to use it in different projects.

mormor1971 said:
Hey you could even make each arduino manage a number of cells where the number of cells would be a function of available ADC pins and I/O pins for controlling mosfets and so on.

This is a possibility but right now I wanna keep the design small and clean. Less BOM cost, smaller pcb.

mormor1971 said:
You would need to come up with a resistor network for lowering the cell voltages to something the ADC pins on the arduino can handle and obviously when you scale the cell voltages down you will loose out on the precision of the ADC's.

Not really if you choose your resistors widely with a little overhead for safety and using a 12bit ADC instead of arduino's 8bit. I used this technique on my FCDS too.
 
I've started doing this using ATTiny85's powered by each pack and then an ESP8266 as the control so to speak. Just monitoring currently but going to use a power resistor for top balancing.
 
Back
Top