BQ40Z50: Learning Cycle Question

Hi Andy,

Please share a bqStudio log of the learning cycle being attempted since this will give us a better insight to why the gauge is not reaching Update Status of 6.

Regarding the chemID, can you please confirm if the GPCCHEM tool was used to get a match?

Andy Hsu said:

1. What settings or parameters need to be changed in order to properly conduct the learning cycle experiment?

The required settings can be found in the document below:

https://www.tij.co.jp/jp/lit/an/slua903/slua903.pdf?ts=1740760115268 

Andy Hsu said:

. According to the learning cycle specification, the term voltage is set at the minimum voltage of the battery, so I set it to 2.6V. However, in actual usage, the discharge never goes below 3.3V if goes below 3.3V it will trigger CUV protection . In this case, for the learning cycle experiment, should the terminal voltage be set according to the 3.35V we actually use, or should it be set based on the minimum battery voltage?

Please set the term voltage to 2.6V in this case.

Regards,

Anthony

Hi Anthony,

Thanks for replying. 

Below is my log file for the learning cycle

6011.LOG.txt

I can confirm we used the GPCCHEM tool to find the chemID。 Here is my report

Chemistry ID selection tool, rev=2.54		
		
Configuration used in present fit:		
ProcessingType = 2		
NumCellSeries = 1		
ElapsedTimeColumn= 7		
VoltageColumn = 8		
CurrentColumn = 9		
TemperatureColumn = 10		
		
Best chemical ID : 7724	Best chemical ID max. deviation, % : 1.79	
		
		
		
Summary of all IDs with max. DOD deviation below 3%		
		
Chem ID	max DOD error, %	Max R deviation, ratio
7724	1.79	0.71
5187	1.82	0.95
7557	1.87	0.46
5929	2	0.67
7028	2.05	0.72
7336	2.15	0.55
2085	2.23	1.21
7670	2.31	0.67
5410	2.33	1.48
7706	2.35	0.99
7434	2.36	0.38
7671	2.43	0.72
7249	2.52	1.09
5936	2.63	0.82
5432	2.65	0.7
7476	2.71	0.81
5413	2.77	0.9
7783	2.77	0.64
2144	2.78	5.68
7574	2.8	0.41
5247	2.85	0.21
7228	2.85	0.91
7174	2.99	0.89
		
Max. deviations for best ID is within recommended range. Chosen best chemical ID is suitable for programming the gauge.		
		
		
Selection of best generic ID for ROM based devices like bq274xx		
		
		
Device / Family #1		
Generic Chem ID	Device/ Voltage/ Chemistry	max DOD error, %
3142	bq27421-G1D: 4.4V LiCoO2	17.73
354	bq27411-G1C: 4.35V LiCoO2	19.27
128	bq27421-G1A: 4.2V LiCoO2	19.79
312	bq27421-G1B: 4.3V LiCoO2	21.76
Best generic ID 3142		
Warning: Generic ID Deviation is so high that it is most likely due to anomaly in the data. Please check that data files have recomended format, units and test schedule		
		
		
Device / Family #2		
Generic Chem ID	Device/ Voltage/ Chemistry	max DOD error, %
1210	bq27621:  (ALT_CHEM1) 4.3V LiCoO2	18.34
1202	bq27621: (default) 4.2V LiCoO2	18.63
354	bq27621:  (ALT_CHEM2) 4.35V LiCoO2	19.27
Best generic ID 1210		
Warning: Generic ID Deviation is so high that it is most likely due to anomaly in the data. Please check that data files have recomended format, units and test schedule		
		
		
Device / Family #3		
Generic Chem ID	Device/ Voltage/ Chemistry	max DOD error, %
3142	bq27426: (ALT-CHEM2) 4.4V LiCoO2	17.73
1202	bq27426: (ALT_CHEM1) 4.2V LiCoO2	18.63
3230	bq27426: (default) 4.35V LiCoO2	20.03
Best generic ID 3142		
Warning: Generic ID Deviation is so high that it is most likely due to anomaly in the data. Please check that data files have recomended format, units and test schedule		
		
		

Why should the term voltage to set at 2.6V not 3.35V? 

I tried modifying the Term Voltage and found that when I changed it from 2.6V to 3.35V, the FCC dropped below 4000mAh. However, when I changed it back to 2.6V, the FCC returned to normal. I would like to know if the Term Voltage is a factor in calculating the FCC.

Best Regards,

Andy

Hi Andy,

Thank you for sending the log file. It seems like the gauge is receiving the correct resistance updates for the Update Status to increase, however not the Qmax update. Can you please confirm why the first discharge goes to ~2.3V when the second is 2.6V?

Andy Hsu said:

Why should the term voltage to set at 2.6V not 3.35V? 

I tried modifying the Term Voltage and found that when I changed it from 2.6V to 3.35V, the FCC dropped below 4000mAh. However, when I changed it back to 2.6V, the FCC returned to normal. I would like to know if the Term Voltage is a factor in calculating the FCC.

The purpose of the learning cycle is for the gauge to understand the full capacity of the cells in use. There are parameter changes that can be made after the learning cycle if you do not want to use the full capacity of the cell, such as the reserve capacity. The FCC value doesn't directly rely on the term voltage, however it does rely on the Qmax, which relys on the amount of passed charge seen during the charge or discharge period. If you are discharging less, then there can be a difference here, however we recommend to disregard the FCC and other SOC values during the learning cycle since the gauge is learning the necessary values to begin taking accurate measurements.

Regards,

Anthony

Hi Anthony,

Thanks for replying.

Anthony said:

Can you please confirm why the first discharge goes to ~2.3V when the second is 2.6V?

I used the electronic load to perform the discharging test. I stopped the electronic load when the voltage dropped to 2.6V, but I'm not sure why it fell to 2.3V. That might be the cause for failing the learning cycle test? Am I right? 

And I would like to know if the 5hrs relaxing is required  or [VOK] and [RDIS] bits are clear and [rest] bit is set than I can move forward without waiting for 5hrs .

Anthony said:

The purpose of the learning cycle is for the gauge to understand the full capacity of the cells in use.

Does the OCV table is determined when the chemstry ID is selected and it is no relevant to learning cycle test?

Best Regards,

Andy

Hi Andy,

Andy Hsu said:

I used the electronic load to perform the discharging test. I stopped the electronic load when the voltage dropped to 2.6V, but I'm not sure why it fell to 2.3V. That might be the cause for failing the learning cycle test? Am I right? 

Potentially, we would need a log file with the change to confirm this.

Andy Hsu said:

And I would like to know if the 5hrs relaxing is required  or [VOK] and [RDIS] bits are clear and [rest] bit is set than I can move forward without waiting for 5hrs .

Yes, if the bits are correct an a Qmax update has been observed the entire rest period is not needed.

Andy Hsu said:

Does the OCV table is determined when the chemstry ID is selected and it is no relevant to learning cycle test?

The OCV table is programmed when the chemID is programmed. This is important to the learning cycle since it is used in the resistance update.

Regards,

Anthony

Hi Antony,

Thanks for replying

Anthony said:

The OCV table is programmed when the chemID is programmed. This is important to the learning cycle since it is used in the resistance update.

As you know, OCV table plays a critical role in deciding the resistance update. 

And OCV is determined by chemistry ID. We need to confirm the correct ID is applied. Can you provide OCV table (voltage vs RSOC) of the chemistry ID 7724

Best Regards,

Andy

Hi Andy,

It is not possible to show a direct correlation between the voltage and RSOC since this is based on the load profile being used. The RSOC is calculated from the RemCap and Full Charge Capacity calculations and is not directly pulled from the OCV table.

Regards,

Anthony

Hi Antony,

I have another problem. When I was in the third step (charging to full) the [FC] flag was set. But after relaxing a short period(less than 2hrs). The [VOK] and [RDIS] were clear and [REST] bit was set. However, the [QMAX] bit wasn't set. 

Since I didn't have this log file, I want to confirm if the relaxing more is require for QMAX update.

Also I can confirm the temperature is between 10~40. The Voltage is outside the flat region. But I am not sure about the delta capacity and offset error. Can you explain more detail. Thanks.

Best Regards,

Andy

Hi Andy,

Can you please confirm that in the relax period between the discharge to empty and charge to full, that the [VOK] becomes cleared and the [REST] bit becomes set?

Regards,

Anthony

Yes, I can confirm this. The [VOK] bit is clear and [REST] bit is set.

Hi Andy,

Understood, thank you for confirming. Would it be possible to share the bqStudio log file of the entire learning cycle? This will allow us to take a deeper look at this time.

Regards,

Anthony

Sorry I don't have this log file for this learning cycle. 

Here is an update in recent learning cycle. The update status goes from 05 to 06 which I think is a successful learning cycle. However, when I tried to modify the update status from 06 to 02, it failed.  But if I click the gauge enable button, it change from 06 to 02. However, if I disable the gauge_en, the Ra Table will not be updated base on the BQ40Z50 TRM. If there any solution to this problem?

Best Regards,

Andy

Hi Andy,

Yes, if the learning cycle reaches 0x06 this means that the learning cycle has been completed. GAUGE_EN should be sent again once the device has been programmed in its final application to enable the gauging functionality.

Regards,

Anthony