• State TI Thinks Resolved
  • Locked Locked
  • Replies 2 replies
  • Answers 1 answer
  • Subscribers 64 subscribers
  • Views 526 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 the BQ35100 in EOS (questions about unsealed access, Ra table, SOH, R Data Seconds)

Gaetan LAURE
Prodigy 30 points
Part Number: BQ35100

Hello, 

We are using the BQ35100 in EOS mode with a lithum primary battery and we have 5 questions :

1) When we perform a "Gauge disable" and the a "Gauge enable" (using the pin 5), we loose the UNSEALED access. Then we have to UNSEAL each time we do a "Gauge enable", is it normal ?

(We don't remove the battery, so the BQ35100 still have power supply beetween the "Gauge disable" and the a "Gauge enable")

2) The chemestry programmation only affect the Ra table or it affect other parameters ?

3) We are planning to use the LSH 20 SAFT (13000 mAh) battery but it's not in the last chemestry tab (version 767). So we will not have the good Ra table.

Does it matter for EOS detection since it's based on relative evolution of the impedance ?

4) Concerning the SOH estimation, how can we have the best results without the good Ra table ? Or how can we determine by ourselve the good Ra table ?

5) What is the influance of the R Data Seconds parameter ? By default it's programmed to 15 seconds, can we shorten that to 1 second for exemple ?

Because for us it's a bit long to stay more than 15 seconds at constent consumption, waiting for "G_Done".

Thanks in advance for your reply

  • 0
    TI__Mastermind 24870 points

    I'm assigning to device expert, expect a response within two days.

  • 0
    TI__Genius 10370 points

    Gaetan,

    1) Yes this is a security feature. Everytime you wake the device if it has been previously sealed will come up sealed again. Ideally there is no reason to unseal the device in the field. If you program in the fresh srec this lockout will be cleared until you issue the seal command again. Setting the GE pin low is effectively removing power from the gauge. This is to ensure the gauge is in the lowest possible power mode when not in use. 

    2) It does affect other area's of the flash. All the area's it does touch are chemistry related. mainly the temperature compensation as well as the OCV. Lithium-thynol chloride cells this is less important since OCV should be 3.6 constantly. 

    3) It is recommended to have a new ChemID done for this cell if it is not in the database. Do you have a TI sales you have been working with? 

    For EOS no this is a relative thing, Ra is used for the SOH estimation. 

    4) Without a chemID. the best action would be to use the gauge to run a full discharge, plot the measuredZ over time. We can use this data to construct a new Ra table that matches your profile.

    To discharge fully it would not be a standard constant discharge, but instead, run the application as intended for a few days to take a few measurements. Follow that with a discharge of around 3-5% capacity and a rest of at least 5 hours. Lastly return to the normal profile for a few more cycles. 

    5) This depends on the setup. The gauge continues to take measurements for this time to help calculate the measuredZ. It feeds the values into an algorithm to produce this value. If you chose to reduce it it is recommended you test with the new value first. The goal would be to get consistent measuredZ values under the same load profile. In my experience reducing this value is not that helpful. Drawing 100's or uA worst case over 15seconds is effectively 0 consumed current if you actually calculate it out. In addition after the Gauge_Stop is issued the sampling rate does change and power goes down vs right after gauge start. 

    I hope this information helps

    Thanks,

    Eric Vos