BQ27426: How do I quickly match the FuelgaugeIC's filtered Fcc to the inserted battery?

Hello jonny,

Thanks for you reply.

Reserve capacity 350mAh, to update the battery charging and discharging scenarios.

Charging scenario: RM<270mAh-->charging-->RM≥320mAh-->stop charging-->resting-->battery self depletion-->RM<270mAh-->charging, and so on.
Discharge scenario: RM≤100mAh --> stop discharging --> equipment power off, FuelgaugeIC power off

1、Based on the above configurations and usage scenarios, will Filtered FCC be updated during use?

2、Does Filtered FCC and True FCC have any limit on the amount of change when updating? For example, the change value for each update can't be more than 100 max or something like that (100 is just an example, it could be some other number).

储备容量350mAh，更新一下电池充电和放电场景，

充电场景：RM<270mAh-->充电-->RM≥320mAh-->停止充电-->静置-->电池自损耗-->RM<270mAh-->充电，依次循环。

放电场景：RM≤100mAh-->停止放电-->设备断电，FuelgaugeIC断电

1、基于以上的配置和使用场景，Filtered FCC会不会在使用过程中更新？

2、Filtered FCC和True FCC在更新时，有没有变动量限制？例如，每次更新的变化值，最大不能超过100之类(100仅是举例，可能是其它数字)。

Hello Jonny,

Sorry, I think I didn't describe my question clearly, the description in the FAQ is not relevant to my question.

Redescribe my question:

Does True Fcc have a limit on the difference between the current value and the previous value when updating? If there is a limit on the difference, then if the current value exceeds the limit compared to the previous value, will True Fcc be updated or discarded?

Is there a limit on the difference between the current value and the previous value when the Filtered FCC is updated? If there is a difference limit, then if the current value exceeds the limit compared to the previous value, will Filtered Fcc be updated or discarded?

抱歉，应该是我的问题没有描述清楚， FAQ 中的描述与我的问题没有相关性。

重新描述一下我的问题：

True Fcc在更新时，本次值和前一次值的差值有没有限制？如果有差值限制，那么本次值与上一次值相比超过了限制值，True Fcc会更新还是被丢弃？

Filtered FCC在更新时，本次值和前一次值的差值有没有限制？如果有差值限制，那么本次值与上一次值相比超过了限制值，Filtered Fcc会更新还是被丢弃？

Hello Jonny

Thanks for you reply.

The following question was copied from another thread, please help answer it, thanks!

1、Temp change --> Is there a time limit for the temperature to exceed 5°C? Within what time period will a temperature change of more than 5°C be determined to satisfy the condition?

According to my understanding of the FAQ, when Smooth is enabled, SOC is calculated using Filtered Fcc, but there is no description of the Filtered FCC update condition：

2, What scenario will be synchronized to Filtered Fcc after True Fcc update?

3, What scenario will Filtered FCC update?

Hello Jonny，

1、If there is no time limit, is the temperature of the last acquisition compared with the temperature of the current acquisition, and is updated only after it exceeds ±5℃?

If the trend of temperature change is 25-->26-->27-->28-->29-->30-->31, is this a case of temperature exceeding 5℃?
If the above counts as more than 5°C, should the temperature be 32°C or 37°C for another update?
Also, how often does the FuelgaugeIC check the temperature?

Jonny said:

There is not a time limit for the temperature to exceed 5°C.

2. In other threads, there are 7 True Fcc update conditions listed, are there any similar update conditions for filtered Fcc?

3、Without sending Smooth_sync command and SMOOTH on by default, will True Fcc be synchronized to Filtered Fcc and if so, under what circumstances will the synchronization be triggered?

4、How is Filtered Fcc filtered? Can you provide the corresponding filtering formula?Or a simple text description of the algorithm roughly?

Best Regards，

wang dong

Hello Jonny，

Thanks for you reply。

1、The initial value of the temperature is 25°C, and any time it is updated to 30°C, it is considered to be more than 5°C. The initial value of the temperature is updated to 30°C, and waiting for the next temperature change to 25°C or 35°C is again judged to be more than 5°C. Is this understanding correct?

Jonny said:

If the temperature changes from 25°C up to 30°C then an update will occur. 

2、Fcc update is predicated on Qmax update, is this understood correctly?

3. Is the following diagram an updated formula for Qmax?

4, Is the formula in question 3 the only Qmax formula?

5, What is the basis for the new battery to be able to update Filtered Fcc during operation?

6、Checking the TRM, I found that there are two limiting parameters for Qmax update, if the updated value of Qmax exceeds the limitation, is it discarded and still use the last Qmax value?

7、Directly inserting a degraded battery, the Fcc update condition is satisfied(Does not perform a complete discharge cycle, but satisfies the conditions of rest, start of charge, start of discharge, etc.), but the Fcc can't be updated, is it because the update value of Qmax exceeds the limit and is discarded, resulting in the Fcc can't be updated?

Best Regards, 

Wang Dong

Hello Jonny.

Thanks for you reply。

Regarding question 7, still don't get it, refine the question:
1、The reason why a new battery inserted into the device can update Qmax and Fcc without a full charge-discharge cycle, my understanding is that the new battery matches the Fs file loaded by FuelGaugeIC, and the updated value of Qmax does not exceed the limitation, and it will be updated in time, thus updating Fcc. Am I correct in my understanding?

2, If the understanding of question 1 is not correct, please help to explain the theoretical basis that a new product battery plugged into the device can be updated without full charge-full discharge?

3、Deteriorated batteries inserted into the device, why must meet the full charge-full discharge cycle?Please help to explain the theoretical basis?

4. According to the Qmax update formula, when a degraded battery is inserted into the device, only charging action (not full) or discharging action (not empty) is needed to update Qmax. why can't Qmax be updated in practice?

5. When a degraded battery is inserted into the device, the reason why Qmax can not be updated, my understanding is: Qmax is recalculated, but the calculated value of Qmax exceeds the limit compared to the previous value, and is discarded, and the previous QMAX value is still used. Is my understanding correct?

Wang Dong said:

7、Directly inserting a degraded battery, the Fcc update condition is satisfied(Does not perform a complete discharge cycle, but satisfies the conditions of rest, start of charge, start of discharge, etc.), but the Fcc can't be updated, is it because the update value of Qmax exceeds the limit and is discarded, resulting in the Fcc can't be updated?

6、Deteriorated battery inserted into the device, Fcc can not be updated, my understanding is: because the Qmax exceeds the limit, using the previous Qmax value, resulting in the Fcc is still the value of the last time. Is my understanding correct?

7, If the above 5 and 6 are not understood correctly, may I ask: what other factors can cause Qmax and Fcc not to be updated?

8, The update of Qmax needs to get DOD2 and DOD1. is there any limit on the amount of change for DOD1 and DOD2 similar to Qmax?

9, Using TI Studio to capture the logs, please help to analyze the reason why Qmax and Fcc are not updated.
Test procedure:
Discharge for 8 minutes and let it sit for 1 hour or more. Can meet Fcc update conditions (Start of discharge and Every hour in relax)

TI_chg_relax_dsg_test_Fcc_1110.zip

Best Regards,

Wang Dong

Hello Wang, 

To answer your first few questions, the gauge requires a true learning cycle (UpdateStatus = 0x03) before the device can gauge the aged cell accurately. Section 6 Gauge Learning Cycle to Improve Gauging Accuracy of the Quick-Start Guide for bq27426 Fuel Gauge provides some insight for the scenarios in which Qmax is most likely to update. Additionally there is not a change limit for DOD1 and DOD2 similar to that of Qmax. 

For your last question, this is a repeat question from this thread. 

Regards, 

Jonny. 

Hello Jonny,

Thanks for you reply.

Sorry for asking a bit too many questions.

We have looked at the above information, but can't get the answers we want,

Adding some context:

The new battery has already performed Gold Learning and loaded the FS file after Gold Learning.

We have a battery usage scenario that does not satisfy full charge, full discharge, and rarely discharges, the discharge path:
a. Battery self-depletion
b, the battery discharged to a certain value, at this time the Gauge will Power Off，and the device Power Off.

So we are not sure whether the Gauge can match with the new product when the battery is gradually degraded during the use of the battery.

Because the product has been shipped, we need to know in detail in order to correct the program in time.

1, In the figure below, you reply Filtered FCC can be updated in our application scenario. Can you provide the theoretical basis?

Note: Our battery usage scenarios do not have full charge, nor full discharge. Most of the time it is in rest mode.

2、If the battery is used in the following scenario, will the Filtered FCC be updated during the long term use of the new battery?

RM<270mAh-->charging-->RM≥320mAh-->stop charging-->Relax-->battery self depletion-->RM<270mAh-->charging, and so on.

3, Can you provide a theoretical basis for the conclusion (Yes or No) of question 2?

Hello Jonny，

Thank you very much for your reply.

SMOOTH EN is ENBALE. We will continue to study the filtered FCC data.

There are several more questions that need to be consulted:

1. Which registers are used to confirm that Gauge has entered Relax mode?

2. Which registers are used to confirm that Guage has successfully Relax and can update DOD2 and Qmax?

3. Do I need to modify Gas Gauging State Update status to 0x03 during use? In our FS file, Update status=0x00.

Hello Wang,

1. & 2. The OCVTAKEN bit can be used to determine if the gauge has entered relaxation. Please see the image below from the Technical Reference Manual. 

3.  Please refer to the image below taken from the Achieving The Successful Learning Cycle document for reference on Update Status. 

Regards, 

Jonny. 

Hello Jonny，

Thank you very much for your reply.

The Achieving The Successful Learning Cycle document has the same process as the Battery Gold Learning.

I may not have explained the background clearly:

We are system side application.
After successfully performing Gold Learning with a new battery, the factory FS file (containing ChemData) is generated.
Device startup process:
Insert new battery -> MCU power up -> load factory FS file -> enter charging or resting scene
Insert degraded battery -> MCU power on -> load factory FS file -> enter charging or resting scene.

The charging or Relax scenario is as follows:

Wang Dong said:

Reserve capacity 350mAh, to update the battery charging and discharging scenarios.

Charging scenario: RM<270mAh-->charging-->RM≥320mAh-->stop charging-->resting-->battery self depletion-->RM<270mAh-->charging, and so on.

Our questions are:
1, Directly plugging into a degraded battery, in our system side application scenario, what is the best way to make Gauge's Filtered FCC match the degraded battery quickly?
2, Inserting a degraded battery directly, in our system side application scenario, what is the method to make the Filtered FCC of Gauge match the degraded battery gradually?
2、Using a new battery, during the natural degradation process, can the Filtered Fcc of Gauge keep matching with the battery in our system side application scenario?

Hello Wang, 

The best way to have the filtered FCC match the newly inserted cell is to run a true learning cycle (UpdateStatus = 0x03) before the device can gauge the degraded/aged cell accurately. 

In the scenario you provide in your last point, please see my previous response.  

Jonny said:

When the SMOOTH_SYNC command is sent then the filtered FCC value will equal the true FCC value.

Regards, 

Jonny. 

Hello Jonny，

Thanks for you reply.

All models of TI's power meters, when applied system side

1、Directly inserted into the aging battery, are there any ICs that can directly match the true capacity of the battery?

2, Directly inserted into the aging battery, do all ICs need to perform the charge --> discharge process to match the true capacity of the battery?

Best Regards，

Wang Dong

Hello Jonny，

Thank you for your reply.

But, you did not answer my question.

Jonny said:

In the scenario you provide in your last point, please see my previous response.  

Jonny Trevino said:

When the SMOOTH_SYNC command is sent then the filtered FCC value will equal the true FCC value.

I'll re-describe the problem:

FuelgaugeIC is used on the system side. The application scenarios are as follows：

Wang Dong said:

We are system side application.
After successfully performing Gold Learning with a new battery, the factory FS file (containing ChemData) is generated.
Device startup process:
Insert new battery -> MCU power up -> load factory FS file -> enter charging or resting scene
Insert degraded battery -> MCU power on -> load factory FS file -> enter charging or resting scene.

The charging or Relax scenario is as follows:

Wang Dong said:

Reserve capacity 350mAh, to update the battery charging and discharging scenarios.

Charging scenario: RM<270mAh-->charging-->RM≥320mAh-->stop charging-->resting-->battery self depletion-->RM<270mAh-->charging, and so on.

1, Inserting a battery that matches the FS file, during continuous use, the battery gradually ages and its capacity decreases, can FuelgaugeIC's Filtered Fcc accurately match the actual capacity of the battery?

2. If yes, can you describe the reason?

3、If not, can you describe the reason?

Best Regards，

Wang Dong

Hello Jonny，

Based on your reply.

My understanding is that all of TI's FuelgaugeICs, in System-side use scenarios, cannot be directly matched to an aging battery when plugged directly into it. A full charge/discharge cycle needs to be performed.

1, Is my understanding correct?

2、Do I need to change the Update status to 0x03 before executing the charge/discharge cycle?

3, How many charge/discharge cycles need to be performed?

Best Regards，

Wang Dong

Hello Wang, 

Wang Dong said:

1, Inserting a battery that matches the FS file, during continuous use, the battery gradually ages and its capacity decreases, can FuelgaugeIC's Filtered Fcc accurately match the actual capacity of the battery?

In the scenario you are describing, if you insert a new battery to the device that matches the FS file, and you were to keep this same battery with the device as the battery degrades over time, then yes, the gauge filtered FCC should match the actual capacity of the battery. To ensure that this value is accurate, it is recommended to charge the battery to full.  

Wang Dong said:

My understanding is that all of TI's FuelgaugeICs, in System-side use scenarios, cannot be directly matched to an aging battery when plugged directly into it. A full charge/discharge cycle needs to be performed.

This understanding is correct. Yes please force update status to 0x03. At least one charge and discharge cycle. 

Regards, 

Jonny. 

Hello Jonny.

Thanks for you reply.

Insert aging battery, FuelGaugeIC needs to match exactly.

1, update status = 0x03, is it necessary?

2、Can I update Qmax and iFiltered Fcc if I only do charge-discharge cycle?

3、If I don't need to match exactly, what do I need to do? For example:

The real capacity is 300mAh, the initial Filtered Fcc predicted by FuelgaugeIC is 850mAh (FS file is made based on a new battery with a reserve capacity of 350mAh, and the FS file is loaded) After performing some operation, the Filtered Fcc changes to 400mAh or near 400mAh.

Best Regards,

Wang Dong

Hello Wang, 

1. Yes this is necessary for the gauge to accurately gauge the degraded cell. 

2. Please refer to section 2.3 Update of Qmax from the Impedance Track Application Note for Qmax updates. 

3. You could use a host to save the last settings of the degraded cell before the cell is removed, then when you reinsert a degraded cell, you use the host to reprogram the saved settings when another degraded cell is inserted. This way the settings in the gauge will correspond to the settings of the degraded cell. 

Regards, 

Jonny. 

Hello Jonny,

Happy Thanksgiving!

Is there a method to update Filtered Fcc to the vicinity of degraded battery capacity without fully charging or discharging?(No need for accurate matching)

Jonny said:

3. You could use a host to save the last settings of the degraded cell before the cell is removed, then when you reinsert a degraded cell, you use the host to reprogram the saved settings when another degraded cell is inserted. This way the settings in the gauge will correspond to the settings of the degraded cell. 

Best Regards!

Wang Dong

Hello Wang, 

The method I gave in the previous response should allow you to match the degraded cell without having to charge and discharge the battery. This depends on whether the cell that was previously inserted into the device was a degraded cell or not. For this method to work you will have to have a degraded cell in the device, then before this cell was removed, a host will have to save the settings. Then when you reinsert a degraded cell into the device, the host will have to reprogram the saved settings to the gauge. Once this is done then the settings on the gauge should reflect the degraded cell. 

Regards, 

Jonny. 

Hello,

What does VCT mean?

Jonny said:

If you want the gauge to run simulations under the conditions listed in the E2E thread you listed, then you should not have SMOOTH on. If SMOOTH is not set then the true FCC and the filtered FCC will always equal each other. If SMOOTH is on then the filtered FCC and true FCC will synchronize when terminate voltage is hit or when VCT is hit. 

Best Regards!