connections for? EcityP Smart S5-13 BMS

Canoe

1 mW
Joined
Aug 12, 2010
Messages
19
I ordered an ECityPower "smart" BMS from BMS Battery on Aug. 17, 2010, with my custom voltages, currents, etc.
It shipped DHL on Aug. 19.
It arrived in Ontario, Canada on Aug. 23.
The BMS and a few extras were appropriately and safely packaged.
There were no hidden fees.

The BMS arrived with the battery wiring harness, with 13 black leads. And there is what is likely a heat probe coming off the board for three inches. The board is a 12s board, and is nice and small at 2.2" x 3.2" (sorry for decimal, calipers where handy).

On the ECityPower site, the photo of the board shows a three pin connector in the upper left of the board. There is no such connector on my board. The only negative seemed to be grossly excessive solder where they added the FETs, but a friend told me that helps with heat management. To attach the power leads, I can either heat the solder up or drill through the solder to the hole in the board and bolt the leads to the board. Under a stereo microscope, we can see that the board was not in a clean environment, with debris and lots of spare solder drops, but all of the connections appear good. From the amount of solder used for various components, it appears that the board was built with the surface mount devices first, and then at a later date, additional components were added by hand. This fits with the customizing available when ordered. There are clamp marks on the power FETS so we know the feet were held to the board contacts when soldered and the FET was heat protected during soldering. It's not a top quality board or build, but my friend assures me that the construction (design unknown) is perfectly adequate to the task. There are numerous testing contacts on the board, and one can see an intent from a probe on each, so it appears that the board did go through QA. There is one place where we can see that something sharp was used to remove trace copper that was a fault between two traces. There are a few jagged piece of copper sticking out from each trace towards the other, but from the testing probe indents, we assume that it did pass QA after this correction. This repair was done by cutting through the surface coating on top of the board & traces, leaving a white canyon line through the coating. This has not been resealed (should it be?). The opaque white of the scratched/cut away surface coating is surprisingly consistent with scratched Tung Oil; curious to know the dielectric properties of Tung Oil.

(~edit: speculation superseded by supplier's connection diagram)

From looking at another of the ECityPower boards, it appears that they like to tie together: battery +, load + and charger +. Then ground both the charger negative the load negative each through their own power FETs to battery negative.
From the board nomenclature it appears that the cell wiring harness has a lead go to the positive of each cell, with an additional lead we guess must go to the bottom cell's negative. On the board, this bottom lead is tied through a small resister to what we think is where the load negative ties into the top of the load's power FETs.
So it appears that the board attachments are:
  • the harness, attached as we assumed above,
  • load negative to board,
  • charger negative to board,
  • battery negative to board.

In the absence of confirmation of the above, I'm rather hesitant to wire it up to the cells.

Does anyone have any experience with the ECityPower Smart S5-13 BMS?
 
FETs are n-channel
doesn't add up putting all Drains to battery negative - must be a different design than the other BMS they make
extra R between cell management harness extra lead and dual FETs' Sources doesn't make sense yet
may be wrong to assume dual parallel FETs are discharge and the lone power FET is for charging, but current wise, it makes sense

not adding up yet

arghhhhhhhhhhhhhhhhhh

ah, for a connection diagram!
 
Canoe said:
ah, for a connection diagram!

BOMARC Services in Wyoming is your friend... they do real good reverse engineering for dirt cheap prices... even cheaper than dirt if you let them resell the results to others.
 
texaspyro said:
BOMARC Services in Wyoming is your friend... they do real good reverse engineering for dirt cheap prices... even cheaper than dirt if you let them resell the results to others.

may have to go there

It appears that the cell harness connects as expected, with RC1 through RC12, starting with RC1 to the positive of the lowest cell and proceeding upwards. The far right odd wire, that traces to R36 to the power FETs' Sources, must go to the lowest cell's negative, the pack negative.

The board has a hole for a power connection at the power FETs' Sources. This must go to Battery(pack) Negative, with R36 in place to ensure that power current doesn't flow through the voltage sensing harness lead, or it may be related to a rail-to-rail small voltage sensing issue. This allows for the Load Negative to go to the power connection at the power FETs' Drains, FETs providing switching and current limiting, with the load grounding out back to Battery Negative through the FETs, with Load Positive receiving Battery Positive.

The addition N-channel power FET has its Drain to the Drain of the other power FETs. There's a power connection on the board to this FET's Source! ???

Where to connect for charging....
 
ECityPower supplied connection jpg

now we're safe
 

Attachments

  • s5-13BMS.jpg
    s5-13BMS.jpg
    37.6 KB · Views: 14,201
The ecity 5-13 BMS uses a O2 Micro chip: OZ890

This is a very interesting chip as it is capable of cascading, real time data output, and is programmable via internal eeprom.

At $25 a board, this could be an outstanding data source of a more 'intelligent' BMS.

There is, however, no data sheet available. I have one of the boards, and have reqested a datasheet, but no response from O2 Micro. Does anyone have any info on this chip?

Thanks
Mark.
 
Hi, I have 2 of these BMS too. I got my hands on the programmer, with all the i2c parameters addresses. I2c should be on pins 3 and 5 of the 12-pin connector on the edge of the board. I don't have a lot of time to experiment, so if you want it just pm me. you can communicate with the program via a simple homemade board that connects to the parallel port (but I haven't tried it yet).
 
My friend has this boards , and he hacked the 12 pin connector , but does anybody have the software , apparently there is only s8 software on bmsbattery
 
i made some comments about the pinout on the O2 micro chip in an older thread when i was looking at the one on the ebikekit.com lifepo4 pack. post up what you guys learn.

http://www.endless-sphere.com/forums/viewtopic.php?f=14&t=10169
 
I have a datasheet. I have the PC control App too, where you can find all the I2C registers and their meaning. Here's the pinout!
 

Attachments

  • oz890.GIF
    oz890.GIF
    18.1 KB · Views: 14,002
Hey, I sent you the program via mail, but I didn't get any reply from you; are you sure it didn't go to the spam folder??
 
Just checked, it's in my "sent" folder, so it must be a problem on your side.... the mail is d s d a v o r i n right?
 
ah , yeah , i am an idiot , i gave you my friends mail .. damm ...it..okay thanks , i ll check with him
 
thanks a lot ..and sorry for akward situation... A mans got to know when to admit hes an idiot ... So here i am ::):):)
 
Hello all.

Does anyboby here tried this software ?

I got a BMS wich uses the OZ890 chip and I'm trying to see its configuration by reading the EEPROM.
The software package is presented and supposed to work with an AardVark interface wich I have but unfortunately the options menu of the software doesn't allow me to choose the AardVark interface, only "Discnnected" or "O2 USB" options are available.

Does anybody have a "solution" ? Or maybe someone can post a log of the communication with the OZ890 sothat I can talk with it using other I2C softwares ?
 
hello all,
I also tried to dump the OZ890. Using an aardvark I2C I/F at 200kbps, some data could be downloaded from slave adress 0x30, but it works only every now and then, and only during a few seconds after a reset of the OZ.

The downloaded bytes # 32 to 4B (starting with #0) contain the actual voltage value readings of the 13 elements, in the order from minus end to plus end, 2 bytes each, little endian, scaled at 10VFS.

Not sure about the meaning of the many other bytes.... Hope this can help someone!

cheers
 
You can find all the register mappings in the files inside "C:\Program files\O2Micro\CowBoy\Configurations\". Look at scan.cfg:

#define "Cell1 (mV)" 0x00 0x0103 [0,1,2;0x32,5,3|0x33,8,0;(0);(0)] "" d
...
#define "Cell13 (mV)" 0x0c 0x0103 [0,1,2;0x4A,5,3|0x4B,8,0;(0);(0)] "" d

8)
 
Back
Top