Multi-Cell Slave-Version 2

Threads relating to the BMS system begun by Peter Perkins

Moderators: GregsGarage, retepsnikrep

GregsGarage
Posts: 870
Joined: Tue Apr 01, 2008 5:27 pm
Location: Galashiels, Scottish Borders
Contact:

Re: Multi-Cell Slave-Version 2

Postby GregsGarage » Wed Mar 21, 2012 8:07 pm

I've added some mounting holes and cleaned up the silkscreen layers. Unless someone sees a serious error in the layout I will order 10 boards for testing.
Attachments
4cellRevA.pdf
(164.03 KiB) Downloaded 643 times
Greg Fordyce

Daewoo Matiz
http://www.evalbum.com/4191

GregsGarage
Posts: 870
Joined: Tue Apr 01, 2008 5:27 pm
Location: Galashiels, Scottish Borders
Contact:

Re: Multi-Cell Slave-Version 2

Postby GregsGarage » Sun Mar 25, 2012 8:18 pm

I've placed an order for 10 boards. Peter, are you going to have any time to work on the new software? I don't think we should try and do the pwm for the load resistor at this point (I am still not sure of the benefit of doing this anyway). The main thing is going to be figuring out how to implement the internal v-ref to measure cell voltage.

EDIT: I will see if I can post a BOM for the board soon.
Greg Fordyce

Daewoo Matiz
http://www.evalbum.com/4191

GregsGarage
Posts: 870
Joined: Tue Apr 01, 2008 5:27 pm
Location: Galashiels, Scottish Borders
Contact:

Bill of Materials

Postby GregsGarage » Sun Mar 25, 2012 9:58 pm

Here's a BOM, part numbers are Farnell and prices are for quantities of 100. I've left C20x off the list, it can be added if we need it.

Code: Select all

4 Cell Slave Version 2 Rev A
part #   price   qty   value      ref
1833888   0.082   4   100n         C10x     
1738662   0.220   4   10R 3W       R10x     
9332650   0.011   4   160R         R20x       
9332804   0.025   4   220R         R30x       
9332383   0.025   4   1K           R40x         
9846662   0.074   4   BC818-40     Q10x           
2001671   0.160   4   LED          D10x         
1971857   0.530   4   PIC12F1822   U10x         
8310759   0.240   8   TCLT1009     U20x/U30x


EDIT: The pcbs were £9.04 for 10 (incl shipping) which brings material costs to around £6.35 for a 4 cell board or £1.60 per cell. 8)
Greg Fordyce

Daewoo Matiz
http://www.evalbum.com/4191

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

Re: Multi-Cell Slave-Version 2

Postby retepsnikrep » Mon Mar 26, 2012 4:30 pm

I'll have a look at the software in next few days if i can.
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.

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

Re: Multi-Cell Slave-Version 2

Postby retepsnikrep » Wed Mar 28, 2012 1:20 am

I have re-done the code for the new slaves from the schematic on here, but until we get some boards up and running with a master I can't test it. I'll do a bit more tinkering as we wait for some boards and try compiling it etc. (Note I don't have a spare made up master)

Greg I could do with 4 boards please to make a pack for my electric bike which i have just ruined the battery on by leaving it switched on for 3 months over winter, sound familiar!!!

For this electric bike pack which will be made with 12-16 A123 20Ah pouch cells I'm tempted to use a simple 12F683 master and a serial 16x2 lcd for a minimal solution with just voltage/temp readings, or even to use the analog slave code from years back where no master is used and the slave and master bus lines just control a charger directly or a buzzer/warning led when in use on the bike. I basically can't be bothered to build a master board.

The attached code is neither tested or checked just FYI at this time.
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.

GregsGarage
Posts: 870
Joined: Tue Apr 01, 2008 5:27 pm
Location: Galashiels, Scottish Borders
Contact:

Re: Multi-Cell Slave-Version 2

Postby GregsGarage » Wed Mar 28, 2012 8:47 am

retepsnikrep wrote:I have re-done the code for the new slaves from the schematic on here, but until we get some boards up and running with a master I can't test it.
That was quick, I'll need to update my pbp to the latest version.

Greg I could do with 4 boards please to make a pack for my electric bike which i have just ruined the battery on by leaving it switched on for 3 months over winter, sound familiar!!!
Ooops. :(
I'll send you 5 boards. I am going to order some bits, do you want me to get you some as well? We should probably make up one board to start with, in case there is a serious error in the layout.

I basically can't be bothered to build a master board.
I was thinking this would be a good opportunity to work on a maximte master. We could knock up a simple interface on a breadboard for the master and slave bus.
Greg Fordyce

Daewoo Matiz
http://www.evalbum.com/4191

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

Re: Multi-Cell Slave-Version 2

Postby retepsnikrep » Wed Mar 28, 2012 11:22 am

Yes please send me the boards and bits or you can make them if you feel brave LOL My eyesight isnt great.

My maximite is still in the bag. I'll leave that to you.

I'll probably work on a cut down 12F683 master or analog setup for my end.
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.

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

Re: Multi-Cell Slave-Version 2

Postby retepsnikrep » Wed Mar 28, 2012 7:34 pm

My first code was wrong I forgot to change the reference voltage calculation and the inverted slave bus input etc so bin it.

I think we might get slightly worse accuracy using the FVR 1.024V reference compared to the 1.235v we had before as the calcuation for the constant yields a smaller number. :?

RefVADC con 52377 'Fixed Ref Voltage Calibration 1.024v * 1023 * 100 / 2 = 52377

RefVADC con 63200 'Fixed Ref Voltage Calibration LM385 1.235v * 1023 * 100 / 2 = 63200

We will have to see.

In some new code below please check I have the pic pins on my little diagram assigned to the right parts of the circuit. Note the config is not right yet if you try to compile it but I'm working on that. :wink:
Attachments
SLAVE_V12_28032012_12F1822.zip
(4.31 KiB) Downloaded 584 times
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.

GregsGarage
Posts: 870
Joined: Tue Apr 01, 2008 5:27 pm
Location: Galashiels, Scottish Borders
Contact:

Re: Multi-Cell Slave-Version 2

Postby GregsGarage » Wed Mar 28, 2012 8:46 pm

Peter, the pic pinouts look fine to me.

I'll order the bits next week, no sign of the pcbs yet. I'll make 2 pcbs for starters, one each.

I like the idea of looking again at the analogue setup. I've just put 4 headway 40160s 16ah cells into one of those portable jump start packs and it could do with a simple bms. Here's one approach. Use the slave bus to cutoff both the controller and battery charger. You give the first slave a signal from a "power" switch and if it's cell voltage is between say 2.3 volts and 3.65 volts it turns on the next slave, etc. The last slave then allows your controller or charger to activate. Use the master bus to give a warning if any cell is not between 2.4 and 3.6 volts. This could be a simple warning light, charger/controller cutback.
Greg Fordyce

Daewoo Matiz
http://www.evalbum.com/4191

GregsGarage
Posts: 870
Joined: Tue Apr 01, 2008 5:27 pm
Location: Galashiels, Scottish Borders
Contact:

Re: Multi-Cell Slave-Version 2

Postby GregsGarage » Thu Mar 29, 2012 9:06 am

Some further thoughts on analogue slaves. One downside is that while it gives you a warning, you don't know which cell gave the warning. We could have each slave log some basic parameters, say min and max voltage. Then you need to get the information out of the slaves. A diagnostic routine which could be activated by a series of pulses on the slave bus would be one way to do this. Once in diagnostic mode the slaves can receive commands in the same way that the digital slaves do, of course you would need a different command set. It could be something like this.

Code: Select all

' Command 01 = Send Cell Voltage on Master Bus         
' Command 02 = Send Max Recorded Cell Voltage on Master Bus                
' Command 03 = Send Min Recorded Cell Voltage on Master Bus         
' Command 04 = Reset Max Recorded Cell Voltage to actual Cell Voltage   
' Command 05 = Reset Min Recorded Cell Voltage to actual Cell Voltage   
' Command 06 = Exit Diagnostic mode      
' Command 07 = Turn On Slave Load for 0.5 seconds (Flash Led)
' Command 08 = Report Slave Software Version Number      


For the diagnostic interface, it could be some sort of usb dongle and a terminal program, something using a pic with some buttons and display or maybe the maximite. You would only use it when you wish to get data out of the slaves, so it doesn't need to be permanently attached to the vehicle.
Greg Fordyce

Daewoo Matiz
http://www.evalbum.com/4191


Return to “BMS thread”

Who is online

Users browsing this forum: No registered users and 3 guests