BQ78350: SOC jumps from 43% to 7% after calibration.

Ben Neubauer

Part Number: BQ78350
Other Parts Discussed in Thread: BQSTUDIO

Hi,

I have a BQ78350-R1 on a 7S6P pack of 18650B cells. After following the learning process and uploading the 6 discharge logs to TI and getting the results and updating the parameters, two of my batteries both jumped from 43% to 7% during discharge. They both changed within 30 seconds of each other and the cells were well balanced after. (Two identical batteries operate in parallel with ideal diodes to power multiple BLDC motors).

I understand this means one of the cells has hit the EDV2 threshold (3501mV), which is not something that gets updated in the calibration process. There was still approximately 40% capacity left in the battery when it hit EDV2 and I continued to use it while it read 0% and carefully monitored the voltage.

What is the proper way to set EDV2? I understand Dod at EDV2 is also important but there are no instructions on how to set it properly. Could someone please share an example? I am using Panasonic 18650B cells, a very common cell so why is the default EDV2 value so far off? At 3501mV, an 18650B cell still has 50% capacity.

Do I need to change EDV2, EDV1 and EDV0 and how do I set them?

Thank you

over 6 years ago

0 Ben Neubauer over 6 years ago

Prodigy 70 points

Also, here is the result of my calibration log processing:

GPC CEDV tool, rev=51		
Configuration used in present fit		
ProcessingType=1		
NumCellSeries=7		
CellTermV=3000		
LearnSOC%=7		
FitMaxSOC%=9		
FitMinSOC%=3		
ChemType=1		
ElapsedTimeColumn=0		
VoltageColumn=1		
TemperatureColumn=2		
CurrentColumn=3		
		
CEDV parameters resulting from the fit. If EDVV bit is set to 1, EMF and EDVR0 have to be multiplied by the number of serial cells when written to data flash		
		
EMF	3437	
EDVC0	174	
EDVC1	0	
EDVR1	665	
EDVR0	67	
EDVT0	2380	
EDVTC	9	
VOC75	27547	
VOC50	26659	
VOC25	26265	
		
		
Recommended SOC deviation tolerance at EDV2 point is  < 5% for low temperature and <3% for room and high temperature		
		
Deviations for this set of parameters are given below for each file		
		
file	SOC error, %	pass
roomtemp_lowrate.csv	0.991916343319915	1
roomtemp_highrate.csv	0.29294095186163	1
hightemp_lowrate.csv	-0.595290226137804	1
hightemp_highrate.csv	0.0698523185164408	1
lowtemp_lowrate.csv	-0.772612994873317	1
lowtemp_highrate.csv	0.164660311395023	1
		
Deviations are within recomended range. CEDV parameters are suitable for programming the gauge

0 Matt Sunna over 6 years ago

TI__Guru* 81035 points

Hi Ben,

Here are a few things to look at:
Did you select the correct ChemID for your battery and program this into the device?
There is an EDV_CMP bit in the data flash. This selects whether to used fixed values or to compute values based on the compensation. I recommend setting this bit high. (See Section 9.1.7 of the Technical Reference Manual)
High current loads pull down the battery voltage so it is important to use similar loads to the application when collecting the CEDV curves. I assume you did this?

You may have already found this document that covers the key steps to setting this up: www.ti.com/.../slua924.pdf
If you still are unable to resolve the issue, it would be good if you can send a BQStudio datalog of the event (click the 'Start Log' burron on the Registers tab. Also, it would be good if you can send your Data Flash settings (click 'Export' on the Data Memory tab to save a .gg file).

Best regards,
Matt

0 Ben Neubauer over 6 years ago in reply to Matt Sunna

Prodigy 70 points

Matt,

Thank you for your response.

When you say the EDV_CMP bit computes values based on the compensation - does that bit need to be set then the battery needs to be cycled multiple times before it computes accurate EDVx values?

My CEDV Gauging Configuration is set to 0x0092 = 0b 0000 0000 1001 0010

From the Tech Ref Guide, I have:

FC_FOR_VDQ(Bit 10) = 0 = VDQset on start of qualified discharge(default)

FCC_LIMIT(Bit 8) = 0 = Disabled(default)

VFLT_EN(Bit 7) = 1 = Enabled

FIXED_EDV0(Bit 5) = 0 = EDV Compensation Used (default)

SC (Bit 4) = 1 = Learning cycle optimized for independent charger

EDV_CMP(Bit 3) = 0 = Disabled(default)

EDV_EXT_CELL(Bit 2) = 0 = Minimum individual cell voltage used as EDV detection reference(default)

CSYNC(Bit 1) = 1 = Synchronized

CCT(Bit 0) = 0 = Use CC % of DesignCapacity(default)

Yes, the CHEM_ID is set to 0x2017 which is Panasonic NCR18650B. There is no information that says what settings the CHEM_ID affects though.

Shouldn't setting CHEM_ID update the EDV values? My EDV values are all still the default values which is the main problem:

EDV2 - 3501

EDV1 - 3385

EDV0 - 3031

I only have logs of the overall voltage (7S) and Absolute SOC. The plot is two batteries discharging simultaneously

0 Matt Sunna over 6 years ago in reply to Ben Neubauer

TI__Guru* 81035 points

Setting EDV_CMP = 1 is important because then it will use your new CEDV coefficients instead of the fixed values. Then the EDV values will be determines as a function of battery capacity, temperature, and discharge load. In the curve above, it is hitting the fixed EDV2 value which confirms that is the cause of the SOC drop.
Some users also set the EDV_EXT_CELL = 1, but you should choose what makes most sense for your battery / system.

Matt

Power management

Power management forum

BQ78350: SOC jumps from 43% to 7% after calibration.