Battery Vehicle Society

Please post any general questions about the BMS project here.

A FAQ will appear here eventually.

I would be grateful for anyone's help on a particular point: published papers on cell balance in Lithium chemistry cells. I have for example read with interest a whitepaper written by Dr Kutkut ("Life cycle testing of series battery strings with individual battery equalisers") documenting the drift out of balance of long strings of AGM cells and the consequent early death of the pack.

Is anyone aware of any similar published evidence for Lithium chemistry cells?

Many thanks

Hi jlgh,

I have moved your question to this general questions topic. You have a good question, I can't think of any papers off the top of my head but if anyone knows of any please post them here.

Hi all,
Thanks to the generosity of Greg and Peter for supplying the pcbs i'm in the process of building the bms for my 48s4p pack of headway 16ah cells. I am currently experiencing a problem with getting the master to talk to 1 slave channel of a 16 cell multi board. I can send commands to the slave (ie command 7 puts the load on for a helf sec) but cannot get the master to acknowledge recipt of any message except command 8 which then tells me i'm running V0! The display flashes "BMS alarms cell one error". Master and slave leds d9 and d7 flash. The load cuts in if i adjust the voltage up and cuts out if i adjust down. I know i'm probably just doing something stupid! Any help much appreciated.

slave ver : 111010 - V08 Beta
master ver : 120111 - V0.22 Alpha

ok , by removing c7 and c9 i have all commands working except command 1. This still returns an error. I also removed vr5 and i can now read the id from the ds1820. Screen is still flashing up bms error on powerup. I have tried 1k and 330r in r8. same results.

Remote diagnostics is pretty difficult and sounds like you are nearly there. Comms must be working and being acknowledged for the commands to work.

Does the report slave version command work?

Is the master set to the correct number of slaves?

Have you checked your boards against my high res photos posted on the other threads?

Does the display cell voltage option work on Master?

Try using the latest version of the master code posted on the software thread and report back.

How many channels have you populated on the 16 cell board?

Are they connected to cells?

If you set the master to one slave but actually have more slaves on the board running then you will get an error as the others slaves will be transmitting data on the master bus in sequence just the master is not allowing for it.

Does the report slave version command work?
Yes

Is the master set to the correct number of slaves?
Yes

Have you checked your boards against my high res photos posted on the other threads?
Yes

Does the display cell voltage option work on Master?
No

Try using the latest version of the master code posted on the software thread and report back.
Running master V22 dated 12/01/11

How many channels have you populated on the 16 cell board?
All channels are populated except for the pics. I have only got a pic in channel one.

Are they connected to cells?
Channel one connected to a bench psu. I will connect up all 16 channels to cells later today hopefully.

I just can't understand why all commands except report voltage are working. Thanks for the help!

edit: Just built up one channel on a 16 cell board. Same results.

I think you need to scope/logic analyse the masterbus

See what is appearing on the master bus when it is running.

With it running you should be getting blocks of three commands.

Load off, measure/send voltage, load on (if reqd)


I don't think i ever tried the system with just one slave.
It could be an issue in the Master software related to that. Try a minimum of two slaves.

Up and running now with 8 slaves. Looks like the problem was with running one on its own. Now to get that I2C temp sensor working .........

Good news.

Re the I2C temp sensor it must be a DS18B20 type.

You need to read the sensor then replace my sensor data with yours and recompile the Master code. I suggest fit two sensors as that is how the software is set up. The temp sensor cable must be 3 core and well screened and screen earthed at Master end only.