Battery Vehicle Society

can you think of any reason that the first cell would report the proper voltage and then every cell after the first one reports a lower voltage than the real value. for example the real cell voltages are pretty close to 3.30 and the first cell reports 3.30 and then all the rest report 2.47 and 2.49. i tried swapping the first and second slave pic with each other, and same result, so it must be something in the master software? i am just setting up a new system and trying to get the last few bugs worked out again

I can't.

Post your slave and master software for examination.

If you can get a logic analyzer on the master bus, you can read the raw data from the slaves which will tell you where your problem is.

i stuck a scope on the slave bus coming out of the master and i'm reading a pulse at 1s intervals, but instead of just a pulse it kicks up and then decays to zero like a capacitor discharging, takes maybe 50ms to decay. (this is with the master board disconnected from the slave board) so maybe it's decaying like that because there is no load on it.

the strange part though is that the master bus is spewing back tons of data, with or without the slave bus connected to the slave board.

if everything is working properly and i scope the master bus (with 6 cells connected) i should see 6 groups of information packed pretty closely together, (one group for each cell) and this should repeat at 1second intervals, right?
as far as i can tell the only thing i've changed on these slave boards is the opto's, i'm using these:

http://pdf1.alldatasheet.net/datasheet- ... PC817.html

i'm using PC817, so they should have a CTR of 50%-600%. it looks like i'm getting a full 5V output on them, so i don't think the CTR is the problem. the only thing that could be of concern is the response time.

even if the response time is poor though, i can't think of any way that this would cause such a weird response on the master bus.

any ideas? i'm totally lost

Post the exact software you are using, some good quality pics of your boards, and some video of what it is doing.

With six cells you should see 3 packets of data on the slave bus & 18 packets of data on the master bus per second.

Remember for each cycle the master sends out

1) Turn loads off and wait for response from each slave
2) Read cell voltage and wait for response from each slave
3) Turn on load if reqd and wait for response from each slave

isaac_alaska1 wrote:i stuck a scope on the slave bus coming out of the master and i'm reading a pulse at 1s intervals, but instead of just a pulse it kicks up and then decays to zero like a capacitor discharging, takes maybe 50ms to decay. (this is with the master board disconnected from the slave board) so maybe it's decaying like that because there is no load on it.

If you want to check the master without connecting it to a slave, use a led and resistor to "load" the slave bus (or a spare opto if you have one).

the strange part though is that the master bus is spewing back tons of data, with or without the slave bus connected to the slave board.

Yes, that is strange. You shouldn't see anything on the master bus until a slave recieves a command on the slave bus. However if a slave interputs some noise on the slave bus as a command, then it will respond on the master bus and send that command to the next slave, etc.

I am sure you know this, but someone reading this won't, so I am posting a useful link regarding oscilloscopes and measuring circuits that don't share common earths/ground which is the case with the master and slave buses.
http://www.eevblog.com/2012/05/18/eevblog-279-how-not-to-blow-up-your-oscilloscope/

thanks greg. i'll try loading it as soon as i get a chance and try to post some pictures. been in the process of moving out of the lab we have been using so it's a slow process.

great video btw, i'm using an isolated supply to power the master and an isolated six cell pack for the cells, so it shouldn't be an issue for me, but i've definitely seen it cause problems. there was a group here that spent a week building a full bridge rectifier, because they kept tripping the mains breaker every time they would connect it. evidently the ground on the scopes we have here is a very solid connection

Hi Everyone,

Hi have been following this topic for some time. Very good work done here!

I would like to build the PCBs however I can not find the original files, only the PDF/JPEG files.
I would like the V2 Master and the 4 Cell SMD version of the slaves, can someone point me into the right direction here?

Thanks

Hi casemod, I'll upload the slave files when I get a chance, busy getting ready for Christmas at the moment. Peter may have a V2 master pcb, I'm sure he'll be along and let you know if he has any.

GregsGarage wrote:Hi casemod, I'll upload the slave files when I get a chance, busy getting ready for Christmas at the moment. Peter may have a V2 master pcb, I'm sure he'll be along and let you know if he has any.

Cheers. He doesnt, I sent him an e-mail asking.

What company did you use to make the PCB's? I am thinking in using this guys http://imall.iteadstudio.com/open-pcb/pcb-prototyping.html.

Ill send you my e-mail by PM.

Thanks