• State TI Thinks Resolved
  • Locked Locked
  • Replies 2 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 393 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

BQ35100: Using BQ35100 for Lithium Primary Cells

Cody McBride
Prodigy 10 points
Part Number: BQ35100


I am using the BQ35100 to estimate remaining charge in lithium primary cell batteries. I will be using two different battery chemistries:

  • Li-MnO2 (DL123)
  • Li-FeS2 (L91)

 

I notice in the datasheet, that Statement-Of-Health (SOH) mode is recommended for the Li-MnO2 chemistry. From my understanding, this operating mode uses a voltage measurement to estimate remaining charge. In looking at the discharge curve for the Li-MnO2 (DL123) battery curve, I can see that it plateaus for a majority of the operating life of the battery. Wouldn’t this be difficult to accurately gauge the remaining capacity of the battery. I have attached a plot of the battery voltage versus the depth of discharge. On the plot, it can be seen that the battery voltage a 5% Depth of Discharge, and the battery voltage at 50% depth of discharge is the same. In this case, would the BQ35100 report both cases as 100% because there has been no change in battery voltage? It may also be worth noting that our devices run near continuously, that is, that there are a minimal amount of ‘wake’ events.

 

For the Li-FeS2 (L91) battery, we are interested in using the accumulator (ACC) mode to determine how much the battery has discharged. The difficulty we find in using this method, is that we cannot guarantee that the battery is a fresh battery every time. It is not uncommon for us to put partially discharged batteries in our systems. A similar problem occurs as the voltage versus discharge curve plateaus for a majority of the curve, making it difficult to estimate charge based on voltage. This use case is similar to how we use the Li-MnO2 (DL123) batteries, in that they run continuously with minimal ‘wake’ events.

 

Simply put, our objective for the BQ35100 is to accurately assess, and report the charge of a battery as a percent (relative to total runtime). Which method do you think will best support this?

 

Thank you in advance,

 

Cody McBride

  • 0
    TI__Mastermind 24870 points

    I have assigned to apps engineer to help you here.

  • 0
    TI__Genius 10370 points

    Cody,

    You are correct SOH mode does rely on the voltage to determine SOH. I think you will find that the voltage does have a minor slope to it even at that high voltage range. i agree the accuracy will be more compromised, but as you get lower and lower the accuracy will increase. This will lead to you reaching 50% a little early then hold there for a while and finish off strong. The biggest need for most applications is end of life detection.  To be clear even in SOH mode we never are able to give an estimated capacity remaining or runtime. We only report an SOH% relative to the characterization we perform on a new cell of the same type. 

    In ACC mode the gauge is only a coulomb counter. If the user replaces the battery there is no way for us to know if it is new or not. This would need to be handled by your host MCU (if possible). I would not recommend this option if you cannot be sure if the battery is new. 

    Thanks,

    Eric Vos