• State Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 3 answers
  • Subscribers 63 subscribers
  • Views 647 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.

BQ34Z100-G1: Jumps in SoC estimation

Gorka Lorea
Expert 1620 points
Part Number: BQ34Z100-G1

Hi,

I am using a BQ34z100-g1 with an 8s2p LiFePO4 batteries. The nominal capacity of the battery is 3200mAh.

I have obtained the golden image of the battery and load it in the bq34z100-g1. 

Now I am doing cycles to the system simulating the real application. 

It can be seen in green the current through the battery when the motor runs. It can be also seen how SoC (in red) decreases more rapidly when the motor is moved and it decreases in a slower way when teh system is in standby. 

However, after the nigth comes (about 6pm: 18h) the motor is not driven any more and the SoC continue decreasing slowly.  

However it can be seen a jump on the SoC around 11.30h-11.40h. (pointed in the graph below)

This SoC jump is a jump of 5% and take place after more than 18 hours of relaxation of the system. 

- Which may be the cause of that jump? 

Best regards, 

  • 0
    TI__Mastermind 42535 points
    During relaxation, an OCV measurement is taken by the gauge. This OCV corresponds to a particular state of charge. So depending on the OCV seen by the gauge, your RSOC can also jump.
  • 0 in reply to Batt
    Expert 1620 points

    Hi Batt,

    Thank you for your reply. 

    I knew what you are telling in your post (During relaxation, an OCV measurement is taken by the gauge. This OCV corresponds to a particular state of charge. So depending on the OCV seen by the gauge, your RSOC can also jump).

    My doubt is: why the time to enter relaxation is so long? (the SoC jump happened 18h after the system went into standby).

    I have done another test to the system. The new test is similar to the previous test but after 12 hours of system in standby we run the motor again a few times during half an hour.

    It can be seen that in this situation no SoC jump happens. 

    So, both tests make me believe that the time to enter relaxation (and then run a capacity simulation) of the system is bigger than 12 hours (as it can be seen that in second test no SoC jump occured).

    Again, I have the same doubt, is not that time too long?

    I have read in Texas IT algorithm documents the following: After a 30-minute relaxation period is passed, the dV/dt <4 µV/s condition is checked. Once it is satisfied, OCV readings are taken.

    I can understand times longer than 30 minutes (because relaxation times of the battery(in this case LiFePO4, wich has specially long time of relaxation)) but this times seems to be too long.

    What explanation can you find to this behaviour of the bq34z100-g1?

    Best regards,

  • 0 in reply to Gorka Lorea
    TI__Mastermind 42535 points
    While the gauge checks for relaxation for an LFP cell, the time taken to relax can be long, even one sample that doesn't meet the dv/dt condition can trigger a longer wait.
  • 0 in reply to Batt
    Expert 1620 points

    Hi Batt,

    So, do you think this time periods are logical? I find it so strange. 

    If so, this means that my system is never going into relaxation because the motor is never going to be more than 12 hours without moving.

    Do you think is logical to need a relaxation time longer than 12 hours if using LFP batteries? I do not know any system with relaxation periods longer than 12 hours...Perhaps the bq34z100-g1 can not works with LFP batteries.

    Best regards,

  • 0 in reply to Gorka Lorea
    TI__Mastermind 42535 points
    After a chg LFP cells can take longer to relax. I think your config has LFP relax set in the pack config registers. That is definitely the cause for a few cells taking longer to relax.
  • 0 in reply to Batt
    Expert 1620 points

    Hi Batt,

    Ok. I suppose you mean the LIFE bit of flagsB high byte. 

    What is exactly the purpose of that bit?

    I can read in the datasheet and in other documents related that LIFE bit Indicates that LiFePO4 RELAX is enabled. But this is not too much information.

    What is the difference between setting this bit o leaving it at 0? 

    Best regards,

  • 0 in reply to Gorka Lorea
    TI__Mastermind 42535 points

    LFP cells take a much longer time to relax and the voltage stabilization rate isn't at par with other Li polymer cells. This enables the IT algorithm feature to take that into account when the cell used is an LFP cell.