Battery Vehicle Society

The allegor current sensors have some pretty solid bus bars for the main current. I crimped and soldered some big ring terminals directly to them for high current tabs. No issues with 100A on my setup.

Thanks, at least I know how to connect it now At least one of the people on my forum has seen instantaneous currents of about 140A on his Cycle Analyst, but he's running his controller a lot higher than mine. I see from the datasheet that the 200B is good for -/+200A anyway, I'm pretty sure I'm not going to exceed that!

The Allegro terminals are actually rated for 4X the max load. So the 200A sensor will handle/pass currents of upto 800A before acting like a fuse.

*Definitely* enough then Got most of the components now, just finding some time to start the build

I have designed my first pcb with kicad, picture posted below. The allegro current sensor isn't rendered properly in the picture as I adapted the part from an existing library part. I have just put 2 6.3mm holes in the board for connecting the power leads to, and a 3 pin header for connecting wires to the master. Should make for a tidy connection when using the Allegro current sensor. Any thoughts before I order some boards. I will post the complete kicad files for the board after a few final tweaks.

Ok, my build is going well - have populated 2 of the 25-cell slaves and completely built the master, programmed all the PICs, but - the damn SV2000 is taking ages to get across the Atlantic so I can't test anything yet!

In the meantime I've created a modified spreadsheet from the bits that I've bought as an updated parts list:

https://spreadsheets.google.com/ccc?key=0ApdAEVb_S04pdDR4N3JYQnpIcUhaWkctYVZXOWNtSUE&hl=en_GB&authkey=CKWt7ZcP#gid=0

It's still a work in progress but most of it is there, complete with Farnell part numbers and prices. Hope it helps anyone else taking on this project!

Board looks good Greg. I assume the yellow things are capacitors? What is the resistor for?

Ian thanks for parts list very useful.

The 2 caps and current limiting resistor are shown in the data sheet's typical circuit for the device. Will try and get the gerbers sent to ITead this weekend and see how they come out.

Greg,

I'd like to share one of my KiCad experiences with you. It is well worth checking the soldermask layers in a gerber viewer because I had an issue with a 3mm led that was in the standard library that you get with KiCad. The pads on the footprint file (viewable from cvpcb and fixable from pcbnew module editor) was set to be visible only in the front soldermask, meaning that soldering from the back side of the board would be impossible It's easily fixed by editing the pads and checking the rear soldermask layer but it's one to watch out for Have a look in your module library and you'll see what I mean. (3mm led).
For the record I preferred the gerber viewer that was part of the GEDA suite, it seemed easier to use.

Derek

Thanks for the tip Derek. I have placed an order with ITead, so fingers crossed I got it all right. I agree with you on the gerber viewer in gEDA, I prefer it over the one in Kicad. Kicad has an advantage in that it is cross platform, gEDA only runs on Linux, but I think both programs will read each others format.

I want to do a new slave board. I am thinking of a small board of 10 x 10 cm, the largest size that ITead does. At $28 for 10 boards the price is very good. To get as many slaves as possible on a board, I am thinking of going for a SMT design. I am working on an unrelated project and want to try reflow soldering using a cheap toaster oven. If that project works out I will set my sights on a small multi-slave board.

The TellyMate cheat sheet is a good idea (Derek posted about this on the pb slave thread). I haven't been having very much luck with cheap 3.5 inch lcds and the SV2000 chip. The bottom of my display is a bit garbaged. I did send my master and display to Peter a while back for him to try on his setup and he confirmed my display was causing the problem. It seems the SV2000 and some of the cheap lcd displays don't play together nicely. The same display does work well with the Tellymate though, which seems to indicate that the Tellymate has a cleaner signal. I probably won't worry about it to much until the next version of the master is available.

On a side note, I have been looking recently at the Parallax Propeller. It is an interesting chip that runs 8 parallel cores (Parallax calls them cogs). Each cog can run an independent task. So one cog can look after cell voltage measurements, another the user interface, etc. Back to what got me started on the Propeller chip, built in video generator. Composite video and even VGA can be directly output by the chip, no Tellymate or SV2000 needed. I know Peter won't like the idea of a completely different chip for the master, after all we have only just got the hang of the pics. But there is a basic complier availible which may aid in porting existing code and some neat products as well. Have a look at the Spin Studio range of boards and the View Port software at this site http://ucontroller.com/index.html.