connections for? EcityP Smart S5-13 BMS

rolf_w said:
great to see new faces in the attempt to master the OZ890 (which still is my preferred BMS chip) and the SmartBMS (which still is the best value for money). I would like to understand a bit better what you did, my russian is a bit rusty :wink:
If a Russian is bad, you can use http://translate.google.com/translate?sl=ru&tl=en&js=n&prev=_t&hl=ru&ie=UTF-8&layout=2&eotf=1&u=http%3A%2F%2Felectrotransport.ru%2Fussr%2Findex.php%2Ftopic%2C10507.0.html
rolf_w said:
us__007 said:
I do cascading 2 SmartBMS. Charge and discharge switches allow you to turn it on 100V. I replaced two of the transistor in the circuit current shunt 100 volt.
you replaced the FET on the high side SmartBMS with resistors?
see Fig. 2 SmartBMS_CurrentSens.jpg
rolf_w said:
us__007 said:
Conclusion SCRL chip zener protected at 58B. in the chain of the legs of the resistor R31 replaced with a face value of 2k 30k.
please explain further
see Fig. 1 oz890_HW-Mode (1)_01.png
rolf_w said:
us__007 said:
EFETC connected to the output optocoupler, the leg is configured to turn off the charge.
you followed the AN#16 from O2Micro? 'EFETC-Out' on high side SmartBMS > optocoupler > 'EFETC-In' on the low side SmartBMS ?
I'm just busy looking for the source I have almost determined.Output EFETC can not serve as a source of signal off the charge. I'll be taking a signal from the point of CG, handle and serve on the transistor optocoupler.
rolf_w said:
us__007 said:
Thanks to a colleague Mevial ​​http://electrotransport.ru site for the monitoring program, the configurator and advice.
looks like an interresting and busy site - what about this 'monitoring configurator' program - is it made for the OZ?
r
Colleague Mevial ​​wrote the console program for Linux to query information from the BMS, download and dump configurator http://mevial.ru/oz/. ty program can be found in the theme of the first link or here http://translate.google.com/translate?hl=ru&sl=ru&tl=en&u=http%3A%2F%2Felectrotransport.ru%2Fussr%2Findex.php%2Ftopic%2C10507.msg156763.html%23msg156763
The program eliminates the need for expensive cable for $ 100 http://www.paintyourdragon.com/?p=43
 
us__007 said:
see Fig. 2 "SmartBMS_CurrentSens.jpg"
Can you explain what this circuit does?
us__007 said:
...Mevial ​​wrote the console program for Linux to query information from the BMS, download and dump configurator http://mevial.ru/oz/. the program can be found in the theme of the first link or here http://translate.google.com/translate?hl=ru&sl=ru&tl=en&u=http%3A%2F%2Felectrotransport.ru%2Fussr%2Findex.php%2Ftopic%2C10507.msg156763.html%23msg156763
This looks very useful! Do i understand correctly that this Browser version (.php) could also be used for programming the SmartBMS? thus you'd only need a browser to programm the SmartBMS?
r
 
rolf_w said:
us__007 said:
see Fig. 2 "SmartBMS_CurrentSens.jpg"
Can you explain what this circuit does?
This scheme measures the current, or rather the voltage drop across the channel of the power transistor.
rolf_w said:
us__007 said:
...Mevial ​​wrote the console program for Linux to query information from the BMS, download and dump configurator http://mevial.ru/oz/. the program can be found in the theme of the first link or here http://translate.google.com/translate?hl=ru&sl=ru&tl=en&u=http%3A%2F%2Felectrotransport.ru%2Fussr%2Findex.php%2Ftopic%2C10507.msg156763.html%23msg156763
This looks very useful! Do i understand correctly that this Browser version (.php) could also be used for programming the SmartBMS? thus you'd only need a browser to programm the SmartBMS?
r
No. version of the browser allows you to change the dump, downloaded from the OZ890.
Download the dump can be written in a special program under Linux.
A link to the Ubuntu boot image with a program I have given above as a link to the forum. Coming soon version allows you to change settings OZ890 directly from the program.
Here's a screenshot.
 

Attachments

  • Screen0401.png
    Screen0401.png
    15.8 KB · Views: 6,011
us__007 said:
This scheme measures the current, or rather the voltage drop across the channel of the power transistor.
so far I understood the circuit (I made the drawing :wink: ). But what is the function of the two transistors you replace? they switch on when the 12V power supply output of the OZ890 goes on and off otherwise. Do they protect the voltage sense inputs of the OZ when the power FET is turned off or damaged?


us__007 said:
...Download the dump can be written in a special program under Linux...Coming soon version allows you to change settings OZ890 directly from the program. Here's a screenshot.

I will eagerly wait for this goody! (could the software also be ported to a windows version?)

r
 
us__007 said:
...Here's a screenshot.
I like this logging feature!! Do you think it could be made compatible with the LogViewer http://www.logview.info/vBulletin/?
r
 
rolf_w said:
us__007 said:
This scheme measures the current, or rather the voltage drop across the channel of the power transistor.
so far I understood the circuit (I made the drawing :wink: ). But what is the function of the two transistors you replace? they switch on when the 12V power supply output of the OZ890 goes on and off otherwise. Do they protect the voltage sense inputs of the OZ when the power FET is turned off or damaged?
Yes. According to this scheme is made current shunt BMS EcityP, transistors protect the current measurement inputs OZ890.
rolf_w said:
us__007 said:
...Download the dump can be written in a special program under Linux...Coming soon version allows you to change settings OZ890 directly from the program. Here's a screenshot.
I will eagerly wait for this goody! (could the software also be ported to a windows version?)
r
Perhaps, soon manifest program under Win. will use the converter USB - I2C.
rolf_w said:
us__007 said:
...Here's a screenshot.
I like this logging feature!! Do you think it could be made compatible with the LogViewer http://www.logview.info/vBulletin/?
r
Мы хотели использовать формат Comtrade,
Thanks for the link, try to understand the file format LogView.
 
Does anyone have the data sheet for the OZ890. The previous link gives an invalid zip file error when trying to extract and the password doesn't help with 7zip

Here's the issue, I was using the ECityPower Smart BMS with 8 cells and everything worked fine, but with a new BMS on a 10S pack, it will not allow the 9th or 10th cells to be calibrated. ...and yes it is set up for 10S in the debug and outputs the voltage for cells 9 & 10 during scans, but I just can't enter a calibration value for anything past cell #8 into the debug program.

Any ideas?


.
 
Chuckles said:
I just can't enter a calibration value for anything past cell #8 into the debug program. Any ideas?

According to the data sheet the OZ890 allows external offset calibration only in software mode
(Offset Calibration in HW-mode: Auto done by OZ890:
...In Hardware Mode, the offset calibration will be done automatically once every 30 minutes and the offset values are stored in the corresponding registers, the normal ADC scan will be paused while executing the calibration...

in SW-mode: Must be executed by Microprocessor:
...In Software Mode, the offset calibration will not be done automatically. But the Microprocessor can
request OZ890 to do offset channel ADC in the reserved time slot...
)

r
 
Does ennybudy know how this 20-40a bms will work with 30a constant on 6s battery ?
i chose this bms just cause i cant find in all china a 6s lifepo4 30a constabt bms that will cut off the volts at 2.5v
 
auraslip said:
Never mind, found it.

I wish it was cheaper. How easy is this to use with the provided software?

Deer auraslip
i saw what you wrote in the forum abaut the programable bms 20-40a
i was wondering if you know how this bms will deel with 30a constant on 6s lifepo4
thhanks

ran riegler
 
ranriegler said:
...will work with 30a constant on 6s battery ?

We use (and abuse) the SmartBMS in various setups (LiFePO, NMC, 8-13s and up to 20s, etc.). The 'larger' variation of it uses 3 power MOSFET (4110) for discharge control, the smaller only 2. We measure up to 5kW battery peak current (with CA) equal to approx. 100A_peak which is handled by these 3 FETs. We often discharge at 30A for longer periods (uphill) without damaging the BMS so far...

regards r
 
Can anyone send me the OZ890 datasheet? Planning to make an iPhone based monitor for my e-bike and need to have info about what their i2c implementation looks like.

Regards
 
Hi, I have just bought two of these BMS'es pre-configured for use with 12S Lipo packs. Some initial testing has shown that charging of a pack works well, but when I tried to discharge a single cell down to about 4.10 volts no balancing was taking place. Is this normal behavior?

Also I have just managed to connect to the OZ890 using a Texas Instruments USB to GPIO adapter. I plan to make a simple GUI application to monitor the cells and hopefully also to change the configuration of the BMS. But I do not have the data sheet, could someone be so kind and send it to me ( helgeo (at) hotmail.com ), please :wink:


When reading the registers using a Texas Instruments GUI I get very low readings for the voltages.
Reading 12 bytes from address 0x32h returns: 00 6B F0 6A E0 6A F8 6A 10 6B C8 6A
If I have done the math correctly this works out to be between 2.73 and 2.74 volts per cell, but the pack is fully charged and should read close to 4.2 volts per cell. Any trick needed to get correct readings?

Cheers,
Helge O
 
OK, I'll respond to my own post :)
Found this datasheet for the OZ8920 which seems to be fairly similar to the OZ890:
http://wenku.baidu.com/view/54d861f0ba0d4a7302763a3b.html

According to that the cell readings are 12 bits and should be multiplied with 2.44mV (5000/2048).
That means that my readings are at about 4.17 volts, which makes sense. I also have to check if there is an offset value to apply. The group1 offset should be applied to the first 4 cells, then Group2 to the next 4 and so on.

Next I have to figure out how to switch between Operation registers and EEPROM registers so I can check and hopefully also modify the various BMS parameters. Interesting stuff.

Does anyone know how I can legally buy my own copy of the datasheet for the OZ890?

Regards, Helge O
 
I have no idea if that linux image from that russion site already made it here - anyways

Ubuntu iso (russian) with I2C tool (<- burn to a CD and run it)
configuration page

The tool is called ozflgui. Haven't fount it or the sources anywhere else on the net - yet.

Would be awesome to find an english version since - as I understand it - you can use any USB-I2C adapter with it (possibly even a Raspberry Pi if crosscompiled to arm).
 
us__007,

the circuit that you sketched for current sense seems quite different from the reference design. I would expect to see some low value shunt resistors (typically less than around 0.0025R) as the current sense, but instead I see 3k3R / 1W resistors. What is going on here? These are not typical current sense resistors. Can anybody explain to me how the current sense works in this design? :?
 
felixfurtak said:
...I would expect to see some low value shunt resistors (typically less than around 0.0025R) as the current sense, but instead I see 3k3R / 1W resistors. What is going on here?

hi felix, the OZ890 can measure low voltages (-0.5V to 3.3V) on the SRP & SRN pins thus it is thought to be used with a shunt resistor (one leg to ground) to measure the small voltage drop proportional to the current:
Shunt_measurement0.png

In the sBMS design, however, the shunt resistor to measure the current I_bat is the drain-source resistance of the power FET (R_ds-on). The clamp diodes D7&D8 and the 3k3R resistors R20 & R24 together with the two transistors N2 & N3 are meant to protect the OZ890 shunt measurement inputs (SRP & SRN) from over voltage in case the power FET is swithed off i.e. the drain-source resistance is becoming high:
View attachment 1

Here a simple spice simulation: the power FET is pulsed and I_bat sweeps from discharging to charging. The protection transistors are not included i.e. are considered here to be always 'ON' (which they are as long as the OZ890 generated 12V supply is on):


The sketch of the sBMS design might contain errors, though. Probably one should check the circuit once more to make sure it is correct and interpret its function. For instance it is not clear to me what FET N2 and R20, if used as drawn, should do: now it parallels R20 to R_ds-on (what for?) and disconnects it when the OZ is powered down (to avoid draining the battery?)
r
 
there are a couple of designs based on the OZ890 out there for instance here, here, here or here. Especially the last one
would be something to be tested (2x OZ890 in hardware mode for 26s :D. Anybody uses it or would know where to buy? r
e_chip_bms.jpg
 
that BMS is arranged as a pair of 13S OZ780 BMSs and each one has it's own shunt resistor. the leads come off for each of the 13S packs, but it has just one charging and one discharging lead. you can see where the B+ of the entire pack comes off that trace from the top of #26 where that black wire comes from.

the ebikekit.com BMS for the 36V10Ah lifepo4 pack from several years ago. when it was the big plastic prismatic cells. that BMS has 13S spots, but the top one, #13, is just jumpered so you could even populate the empty transistor and resistor spots with some surface mount parts to make it 13S and then reprogram it but i bet the current program would run 13S lifepo4.

or if it could be reprogrammed for the 12S lipo and use the output mosfets and the charging mosfets to control some other high current contactors or mosfets. the two little mosfets in there would not handle the current or the charging currents for lipo packs.

that is the only BMS i have seen with the OZ790.
 
Hi, I buy this bms.
it seem to work fine.
I have a little problem with balancing mode.
I eneble balancing in idle mode
I set the starting voltage and the difference mV between cells.
It works, but if one cells goes OverVoltage it stop balancing. It's a big problem because if the cell doesn't come back under OV release voltage, it doesn't balance the cells.
Does it work in this way or is there something else to set up?
Thankx to all.
 
If a cell isn't coming back under the OV limit (HVC) then it isnt' balancing anyway, assuming it is intended to drain the high cells in some way when doing that.
 
signo82 said:
I buy this bms.
where did you buy it and when?
... if one cells goes OverVoltage it stop balancing. It's a big problem because if the cell doesn't come back under OV release voltage, it doesn't balance the cells.
Your observation is correct - the OZ890 bleeds only if:
  • The bleeding functions, charge bleeding and idle bleeding (optional), are enabled
  • Battery pack is in charge state (current larger than the charge state threshold) or in idle state (current smaller than the charge state threshold and larger than the discharge state threshold).
  • The highest cell voltage exceeds the bleeding start voltage threshold
  • The lowest cell voltages’ difference exceeds the bleeding accuracy threshold
  • No error event, like OT, UT, OV, UV, OC, SC, is present. If any error event happens, bleeding stops right away and resumes only after all errors are reset.

In the OV case this is bad news since bleeding is inhibited although would be required to reduce the highest cell voltage and balance the pack. So what to do? My strategy is (assuming a higher cell voltage means a higher SOC):

To reduce the charging current to the bleeding current before OV occurs but after the knee where cell voltage starts to rise rapidely with increasing SOC. Thus adopt a typical CC/CV charging strategy i.e. continue charging and bleeding in CV mode until a balanced state is reached.
Unfortunately without communication between charger and BMS this approach only works if all cell voltages are always similar i.e. the pack is already quite well balanced (equal capacities, Ri, SOC, ...). In this case the charger and not the sBMS should stop charging (change from CC to CV mode). Therfore, adjust the sBMS OV threshold above the CC /CV threshold of the charger ; e.g. in a 13s pack OV=4.2V and HVC=13*4.15V=53.95V thus one cell could 'run away' only 54mV from an else perfectly balanced pack before triggering an OV event - not a big margin.

There will always be one cell which reaches OV first e.g. if capacities divert because cells get older. Here I do not have a receipe besides probably changing weak cells and/or helping with an external balance charger. The following strategy will not work: charge with high current and start bleeding at low voltage to max. bleeding time i.e. max. SOC equalizing. Make sure the charger stays in CC mode until the sBMS cuts off due to an OV event. Make sure the cell voltage drop occuring due to the turned off charging current is higher than the OV release condition (OV_thr - OV_rel). Charging and bleeding will restart and produce a 'hickup' mode. However, this could push the highest cell continuously higher into the dangour zone beyond OV.

I don't see a way yet how to rectify this OZ890 flaw. r
 
I bought it from bmsbattery.com few weeks ago.
Tomorrow I'll test it.
I have 12s lipo pack
I set it in this way:
Start bleeding 4050mV
Charger end voltage 4,15*12 = 49,8V
OV = 4170mV
Release OV = 4169mV
so I hope when a cells goes OV, it goes down for at least 1mV
Yes, in this way I could have a continuos on/off of the charger, but I hope it will be only a rare case keeping the charger end voltage at 4,15 per cell.

Thank you for your help.
I'll let you know.
 
Back
Top