BQ27531-G1: SOC and SOH change

Thanks for reply.

Do you have any way SOC and SOH aren't update in the interval?

Just curious what the reason is that you would want to disable updating this information? Updates can not be disabled but you can smooth them out so large jumps are less likely to occur by setting the OpConfig C [SmoothEn] bit.

Thanks,

Jackson

If the battery is left overnight with 50% charge remaining, it may reach 10% the next morning. Remain Capacity and SOH have dropped significantly.

Do you think the gauge IC will improve if gauge don't put it in Relax mode?

Is it right if current thresh is setted 0mA, gauge never put in relax mode?

This does not sound like expected behavior. You should not see an SOH jump that large from simply being in relax mode. I don't think relax mode is the culprit here but rather something else that is triggering an incorrect value for Remaining Capacity. Have you programmed in the correct ChemID and also done Learning Cycles to ensure that the gauge is fully learned before you observe this behavior?

Thanks,

Jackson

>Have you programmed in the correct ChemID and also done Learning Cycles to ensure that the gauge is fully learned before you observe this behavior?

Yes.

When the learned battery is detached and reattached at around 50%, the remaining amount jumps to around 10%.

I am in trouble because the same behavior occurs even though the battery is not inserted or removed.

There are many factors that could be causing an SOC/SOH jump like this. Could you log this behavior and send a log file my way so I can see exactly what is happening internally to the gauge when this jump occurs?

Thanks,

Jackson

I'm sorry that it cannot be get because IC buit-in.

 Please tell me from a different point of view.

Is there any OCV measurement during the interbal?

Even after learning, if I use the OCV command at around 50%, SOC jamp.

For test, I want to stop the impedance track of the gauge IC once and check the issue.

Please tell me how to set it.

Hi,

If an OCV measurement is forced before the cell is fully relaxed, the predicted SOC may be way off due to IR loss. This effect is magnified around 50% SOC where the cell is in the flat region of its OCV curve.

Also, I am not exactly sure what you mean by "IC built in". if you attach an EV2400 to the SCL and SDA lines and use BQStudio you should be able to fully communicate with the gauge and log parameters when this behavior occurs. If the SCL and SDA lines are not availible, can you describe how you are communicating with the gauge?

Thanks,

Jackson