BMS forum.

[mrbigh]

The idea to form this forums is great, it is the right path to start
getting people motivated to do something useful in the electronic
engineering field for free and to overflow this pages with silly
questions mostly of the time, but it is an open source to
everyone.
The idea of the Lipo BMS is already obtainable as finish products from
several suppliers and at very low cost with surface mounted components
and some of it of the size of a dime and at $5.00 in quantities.
If we where able to use such made technology and implemented in a
microprocessor environment for managerial services, I could said we
will be in the wining path.
On my own research and development for the BMS for the NiMH battery
banks I'm working with for my next stage of the PHEV conversion, I
couldn't be so lucky in the use of such premade modules, I developed
my own based on 5 years old technologies IC, soon to be discontinued
from manufacturing because of "obsolete" but there are not to many
choices for this chemistry of Batteries.
I wish good luck to the forum members and I will contribute as much as
possible.

mrbigh
owner of DUAL PWR-PHEV

[Chris Ewert]

mrbigh wrote:
> The idea of the Lipo BMS is already obtainable as finish products from
> several suppliers and at very low cost with surface mounted components
> and some of it of the size of a dime and at $5.00 in quantities.
> If we where able to use such made technology and implemented in a
> microprocessor environment for managerial services, I could said we
> will be in the wining path.
>

If we can find anything that we could integrate, that would be
wonderful, but I haven't come across any yet.

This isn't commercial, but here is what we've found so far:
http://www.edn.com/article/CA6470829.html

Its a design published by Electronics Design Strategy News, which looks
like a fairly good place to start. Its a basic bypass regulator, with a
PIC sensing the cell voltage, and shunting the excess across the cell.
It communicates on a bus with a supervisor. To me, this seems like a
good general strategy. The individual boards can take care of
themselves, but they can also work with the supervisor to balance the
pack and then the supervisor knows each cell's voltage, etc.

Does this seem like a good overall strategy? (see:
http://freebms.googlegroups.com/web/BMS_Overview.jpg and
http://freebms.googlegroups.com/web/BMS_Regulator_Overview.jpg)

If that looks good, a good place to start is probably figuring out what
communication bus we can use. It needs to be optically isolated (I
believe) and bi-directional. It also has to be able to withstand
amazing amounts of EMI. Does anyone have suggestions for a bus type?
I've heard a lot about 1-wire, but I've heard its not necessarily great
for EMI?

Chris

[Andrew Ewert]

On Dec 26, 8:15 pm, Chris Ewert <ch...@infolaunch.com> wrote:
> Does anyone have suggestions for a bus type?
> I've heard a lot about 1-wire, but I've heard its not necessarily great
> for EMI?
>
> Chris

1-wire would seem to be an ideal solution from a software point of
view. There are hundreds of existing projects that use 1-wire for
inter-PIC communication. I'm not sure how the 1-wire approach holds up
under intense EMI however. Does anyone have any ideas to shield 1-wire
from EMI?

Andrew

[HarveyEV]

The 1-wire solution is a pretty common approach for the few existing
BMSs that I have experienced. I will make it a point to ask
these manufacturers that I have purchased systems from for their input
on the effects of EMI on their systems. I have implemented two of
these,
one recently with TS cells with a central control unit and cell
modules design specifically for TS cells. The other was single central
unit with
a 1-wire config to AGM cells monitoring the same set of parameters.

For more info on the TS unit, see http://www.ev-power.com.au/-Battery-Management-System-.html.
The other is a beta unit from EVSource, that currently has no online
documentation, but is a reference to another approach similar to the
Reap Systems (http://www.reapsystems.co.uk/products_bms.html) design
for LiIon cells (specifically TS cells) that do not have individual
cell modules.

Lastly, there is an older, not-so-active EVBMS group on yahoo that may
have a some pertinent info related to the objective of this group at
http://autos.groups.yahoo.com/group/EVBMS/

> Adrew

[mrbigh]

Another alternative for data communication is EVILBus, it is a proven
electrical system developed quite a few years ago, 2 way
communications, optoisolated, scalable, can RxTx almost any protocol
and economic to build. I think originally designed for EV's and
adopted in the Tango, electric vehicle.

On Dec 26, 11:26 pm, Andrew Ewert <Andrew.J.Ew...@gmail.com> wrote:

[perh...@gmail.com]

On 28 Des 2007, 16:26, mrbigh <mrb...@gmail.com> wrote:
> Another alternative for data communication is EVILBus, it is a proven
> electrical system developed quite a few years ago, 2 way
> communications, optoisolated, scalable, can RxTx almost any protocol
> and economic to build. I think originally designed for EV's and
> adopted in the Tango, electric vehicle.

The EVILBus spec looks interesting, with a high level protocol made
for BMS. The spec for EVILBus can be found here:
http://www.casadelgato.com/EV/EVILbusSpecification.pdf

Per

[mrbigh]

Doing some research in my archives I found again the link for the
EVILBus protocol,
http://alamedasystems.com/protocol.html
why to reinvent the wheel again?
mrbigh