Yet another battery management/monitoring system....

Chat about all things battery in here.
Need to know what type to use or size or capacity then again place your thoughts here
bobc
Posts: 107
Joined: Thu Dec 20, 2007 1:11 pm
Location: Knutsford

Yet another battery management/monitoring system....

Postby bobc » Sat Aug 23, 2008 11:01 pm

Thought I'd post a picture of the battery charging/ monitoring system I'm putting together for the kitten project. These little boards are installed across each 12V lead acid battery in the car. Each communicates with the next lower cell in the string, which appends its own data to the message and passes it on - so only a single UART on the main vehicle controller is used & no opto isolation is required. So each board ends up small & cheap (these were about £7 each but they'd be £2 in quantity) I'm using an AVR micro & a big shunt mosfet. The mains battery charger module will be a controlled current source at up to 10A going through the cell string. The micro on this board measures cell voltage, shunt current and there's a temperature sensor near the mosfet.
Image

User avatar
retepsnikrep
Posts: 1387
Joined: Sat May 26, 2007 4:50 pm
Location: North Yorkshire England
Contact:

Postby retepsnikrep » Sun Aug 24, 2008 4:33 am

Very nice Bob 8)

Does it actually work in the car in all the electrical noise? That's the bit worrying me most about my own design.

Does it have a display unit/master?

Will you be releasing it as a freebie?

Can we have a proper look at your schematic and code please, might get some ideas :wink:

Regards

Peter
Regards Peter

Two MK1 Honda Insight's. One running 20ah A123 Lithium pack. One 8ah BetterBattery Nimh pack.
One HCH1 Civic Hybrid running 60ah A123 Lithium pack.

bobc
Posts: 107
Joined: Thu Dec 20, 2007 1:11 pm
Location: Knutsford

Postby bobc » Sun Aug 24, 2008 12:58 pm

Hi Peter,
I'm not worried about electrical noise, it's a small board with a complete ground plane on the back, a big C on all the analog inputs & local V regulator - it will be just fine ;^) Those single chip micros are pretty bombproof actually, give 'em 3.3V & they don't have much option except to do the business....
My plan is to have the overall charging control done by the main vehicle inverter controller, though it would be possible to build a simpler separate charger controller for other applications (it could be integrated in the charger power supply for example). Each board has 3 LEDs of various colours for battery status during charging & my plan was that the main car dash would give a cursory display of any worries (e.g. a low voltage cell or one getting filled too quick & the user would then look at the LEDs on each battery for more info) rather than having a big screen full of battery details.
Code doesn't exist yet - next few weeks when it's rainy I will start to build up the functionality of the board & do any necessary hardware debug while the software evolves. I'll be writing in C using WinAVR, it all looks pretty reasonable. Oh yeah - I still have to buy myself a dragon debugger/emulator - the 6 pin header is the programming connector & the debugwire "ICE" connection is one of the others. I'll post a schematic once it's working !
Bob

User avatar
dillond666
Posts: 139
Joined: Mon May 26, 2008 7:52 pm
Location: Central Scotland

load resistor

Postby dillond666 » Wed Aug 27, 2008 6:25 pm

Hi Bob,
Most excellent work. What is the value and rating of your load resistor and where do you get them? it looks like a neat solution.

Derek

bobc
Posts: 107
Joined: Thu Dec 20, 2007 1:11 pm
Location: Knutsford

Postby bobc » Wed Aug 27, 2008 8:39 pm

No load resistor - all the dissipation is in the MOSFET. In the final assembly there will be an approx 1.5inch cube heatsink on each one with a single fan blowing through all of 'em. I'm calculating on about 30W max for each board (though they will transiently sink up to 120W) Basically if any one board is shunting over 2A the main power supply(charger) for the string is throttled back. While this will increase the time spent in the float charge region (as opposed to the bulk charge region) it will improve overall charging efficiency. And it should generally all be OK as long as the cells are reasonably matched.
It all works great in my head - time will tell how it goes in real life!
Might as well describe the charging strategy:-
First the charger whacks 10A through the whole string. When the cell voltage on any board exceeds 14.4V, it's shunt starts to conduct to hold the current at that level. Soon all the cells should be at 14.4V and the main charger is throttled back to hold the smallest shunt current in the string of cells to just over 0A. Alternatively if any board is over 2A or overheating the charger is throttled back some more. Each board will switch down to a "trickle" charge voltage of 13.7V when the charging current in its associated cell reduces below 0.5A. Eventually (by morning...) all the cells should be in trickle mode, or pretty damn full; then I drive off to work. With the charger unplugged the cell voltage will reduce to 12.7 or so therefore the shunt will be "off". So the code on this board is pretty simple & fail safe, mostly concerned with serial comms for the string monitoring data.


Return to “All things battery related”

Who is online

Users browsing this forum: No registered users and 29 guests