• State TI Thinks Resolved
  • Locked Locked
  • Replies 32 replies
  • Answers 14 answers
  • Subscribers 66 subscribers
  • Views 2845 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.

Original question:

BQ35100: What is the current consumption of BQ35100 in ACC mode? In spec, it is 130uA. But we test bq35100EVM, in ACC mode...

BQ35100: BQ35100 can read the current and charge accumulation values in bqStudio but can't do calibration.

user5874521
Intellectual 450 points
Part Number: BQ35100
Other Parts Discussed in Thread: BQSTUDIO,

I can use bqStudio to read current and other values got by BQ35100.

But I can't do calibration. Though bqStudio don't notice a error, but calibration is failed. We can't calibrate 208mA to 220mA.

While doing calibration, the current will down to 0 mA. And GA will be low, which means BQ35100 stopped working.(Just like doing GAUGE_STOP).

And in bq35100EVM-795 Evaluation Module, When calibrate successfully, the screen is like this.

What's more, in my board, I have to always press UNSEAL, or bqStudio will report an error in following picture. But in  bq35100EVM-795, I don't need to always press unseal.

I have checked my Schematic, the Schematic is the same as  bq35100EVM-795. 

What may cause this problem? It is so strange that BQ35100 can be read but can't be written. Please help!

  • 0
    TI__Guru 52115 points

    We will take a look and get back to you.  Note that this Tuesdays and Wednesdays are public holidays in the US.  Our response will be delayed.

  • 0 in reply to Andy Liu (BMS)
    Intellectual 450 points

    Any update? It is a quite urgent issue in our project. Please reply as soon as possible, thank you!

  • 0 in reply to user5874521
    TI__Mastermind 24870 points

    Hello User,

    Can you stop scanning or make sure you are not scanning the page?

  • 0 in reply to Kang Kang
    Intellectual 450 points

    Thank you for your reply!

    Yes, I'm sure I have stopped scanning and clicked unsealed before doing calibration.

    It is so strange that I can calibrate the BQ35100_EVM. But in my board, I can only read data from BQ35100 but can't do calibration.

    .

    My calibration step is:

    1. Press GAUGE_START

    2. Press Scan to confirm if BQ35100 is measuring current.

    3. Press UNSEAL

    4. Stop Scan

    5. Do calibraiton.

    For example, I can set 50mA to 100mA. And the software will not show any error. But when I measure current again. The current value is still 50mA not 100mA, which means the calibration is failed.

  • 0 in reply to user5874521
    TI__Guru 52115 points

    As far as I can remember, I have not seen that issue with my EVM board.  I will double check today.  

  • 0 in reply to user5874521
    TI__Guru 52115 points

    I have checked my BQ35100EVM board and not seen the same issue. I can calibrate temperature, current, and voltage without any issue.

    The reason why the current appears as 0 mA in the Calibration Window is probably that you disable the Auto-Refresh of the DashBoard. 

    I would suggest you keep the Auto-Refresh on and try again.  

    Andy

  • 0 in reply to Andy Liu (BMS)
    Intellectual 450 points

    Thank you for your reply!

    How to set Auto-Refresh?

    I can calibrate the BQ35100_EVM with software BatteryManagementStudio but I can't calibrate my product. I can only read value on my product's BQ35100.

    I consider the document of BQ35100 is not complete. 

    I even can't find the register address in BQ35100_TRM and BQ35100_CN for setting ACC mode (Register GESEL[1:0]).

    And how to calibrate by Advanced Comm? On the EVM. I write value as the first picture. But when I read. The value is showed as the second picture. Do I miss some step? Before writing flash, I have set unsealed and Gauge_Start. and calibration Enable. Do you have any other document about BQ35100?

  • 0 in reply to user5874521
    TI__Guru 52115 points

    I suggest you create a golden image using bqStudio and BQ35100EVM, and then use the golden image to program the gauge device in your product.

  • 0 in reply to Andy Liu (BMS)
    Intellectual 450 points

    Thank you for your reply!

    1. I can't do calibration on bq35100_evm by Advanced comm though I followed the spec. I can't use golden image to program the gauge device in my product. I can't write to gauge device on my product.

    I can do calibration on EVM, too. But on my product, when I used bqStudio to do calibration, bqStudio didn't show any error but just can't do calibration neither voltage nor current. For example, when bqStudio show 5mA. I calibrate it to 10mA. On EVM. the value can be showed as 10mA. But on my board, the value is still 5mA.

    It is quite urgent and I really need you help. As you say. may be I should set auto-refresh.

    2. Where is auto-refresh, I can only find refresh on bqStudio. 

    As shown in following picture, In slua904, The first step is Press UNSEAL. The second step is Press RESET. But I find if I Press RESET, BQ35100 will be back to sealed.

    Why? is it an error on the document?


  • 0 in reply to user5874521
    TI__Genius 10370 points

    ,

    1) If you have ever sealed one of our gauges every time after you reset or POR the device it will recover back in sealed mode. When you are sealed you will be unable to modify the dataflash which is why calibration will fail. 

    2) When performing calibration the gauge must be in ACC mode [GMSEL1:0] = [0,0]. In SOH mode [1,0] current is not measured, and in EOS mode the CC does not take full 32mS samples. 

    To make things easier please download the base v1.02 srec from the web and program it into the device. Perform calibration first then transition into the correct mode. 

    Thanks,

    Eric Vos

  • 0 in reply to Vos
    Prodigy 110 points

    Hi Eric,

    I want write DF with I2C command according "slua790", the step like below picture indicated, but my operation is not efficient, can you verify in your board and give me a standard operation step?

    The initiation data is 0x80, it's no change after operation according above step.

  • 0 in reply to user5994982
    TI__Mastermind 24870 points

    Hello User,

    This is the proper operation, make sure you are changing and computing the checksum correctly.

  • 0 in reply to Kang Kang
    Prodigy 110 points

    Follow the sample step , i try to write DF address 0x4000(CC Gain), the step like above indicated. Obviously, the write is failed, i don't know why.

    Before i write flash, i have enable calibration,set unsealed, start gague, the control register value like below.

    The checksum calculate step : 0x40 + 0x90 + 0x5d + 0xc6 + 0x73 = 0x266. The last byte 0x66 reverse is 0x99, so write 0x99 to 0x60.

    The data length is 4, so write 08 to 0x61. 

    Have any error that i don't aware? Please give me some advice, thanks.

  • 0 in reply to Vos
    Intellectual 450 points

    Thank you for your support!

    I have another question. Spec said, the current showed by bq35100 has subtracted current consumed by the device.

    What is the default value of EOS Gauge Load Current? Is it 0.35mA？

    Does it means, the total discharge showed in bqstudio has also subtracted the consumption of bq35100 ?

  • 0 in reply to Vos
    Intellectual 450 points

    Hi Eric，

    I find that when the environment temperature is about 60℃ or higher, I pull-up GE (by I2C), then Gauge_start (by I2C), the measure current will be 0mA. But if I pull up GE, Reset BQ35100(by I2C) before Gauge start, BQ35100 can measure current normally. Is it normal or a bug?

  • 0 in reply to user5874521
    TI__Genius 10370 points

    I am not sure what questions are still open on this thread so if i miss something i am sorry. 

    1) In order to update flash your voltage must be above Flash Voltage OK threshold in DF. You must also be unsealed.  In one image above you are sealed and i cannot see the voltage in any of them

    2) The gauge load is an average while active to best approximate the gauge in all modes. Since the gauge power does not flow through the sense R it needs t be accounted for. 

    3) Temp being over 60degC should not impact the ability to measure current or change anything about the procedure. There will be temperature based error such and ADC drift, senseR PPM drift.... ect. Something different must be happening here. Pleas e measure the voltage drop across the sense to see what true current is and make sure GA is active. Send a log file if you see the issue persist. 

    Thanks,

    Eric Vos

  • 0 in reply to Vos
    Intellectual 450 points

    Hi Eric，

    Thank you for your support!

    For your second answer.

    Spec said，bq35100 has subtract a value to subtract the current consumption of itself. Is it a default setting? What is the default value, is it 0.35mA? And how to cancel this setting?

  • 0 in reply to user5874521
    TI__Mastermind 24870 points

    Hello User,

    It cannot be cancelled. This is the power consumption of the device.

  • 0 in reply to Kang Kang
    Intellectual 450 points

    Hi Kangkang,

    I need to use bq35100 to measure the whole consumption of the product including the consumption of bq35100. But bq35100 has subtract the consumption of itself. We want to add this value to improve the accuracy.

     What is this default value? Is it 0.35mA? Or others?

  • 0 in reply to user5874521
    TI__Mastermind 24870 points

    Hello User,

    Yes, we've had questions on this in the past, but unfortunately it is tough to answer. You could try grounding the gas gauge on the Pack- side of the sense resistor. However, the power consumption of the gauge may be too small for the gauge to measure.

    There is no max spec for the bq35100 power consumption due to the fact this part has firmware and flash writes.

  • 0 in reply to user5874521
    TI__Mastermind 24870 points

    In most cases, 0.35 mA is an average. I suggest looking at the power consumption across time in your system to see if you are able to characterize the power consumption in system.

  • 0 in reply to Kang Kang
    Intellectual 450 points

    Hi，

    From the circuit, I think BQ35100 has measured the current of the whole system. But spec said, bq35100 will subtract the current consumption of itself. You say, this setting can't be cancel. I think the subtracted value is setted by TI. There should be a default fixed value setted in the firmware to substracted. Isn't it? Or The current consumption of bq35100 itself is not a fixed value？This value is real measured value by bq35100 itself?

  • 0 in reply to Kang Kang
    Intellectual 450 points

    Hi KangKang,

    Sorry，maybe I didn't express clearly.

    In the BQ35100_TRM, it said, "To subtract this value from the reported current, a value programmed in EOS Gauge Load Current is subtracted for improved accuracy." 

    But I want to use bq35100 to measure the consumption of the whole system.   So I think I should add EOS gauge load current back. I need the value of “Actual Measured (SRP-SRN) Current”.

    Is the default value 0.35mA?

    Is this 0.35mA has already been written in every chip of BQ35100, which means the measured value measured by BQ35100 has already subtract 0.35mA?

    In addition, I can't fin EOS guage load current in CLASS EOS data, Where is it? What is its address?

  • 0 in reply to Kang Kang
    Prodigy 110 points

    Hi kang kang,

    I have ensure more, but still failed. Just like the picture indicated, the gauge is unsealed, calibration status is enable, gauge start is OK, the voltage over "Flash Update OK Voltage" too.

    Please give me some advise, thanks.

  • 0 in reply to user5874521
    Prodigy 110 points

    HI kang kang, 

    Gauge start fail after pull down-up GE pin when temperature over 62℃ . The log fragement like followed picture, full log like the attachment.

    Obviously, The Control status still 0x2080 after we Gauge start 10 times.The Gauge start efficient that Control status transfer 0x2081 after send reset command .

    Through full log we no found the same appearance when temperature is low, so we think this is related with temperture.

    Otherwise, we measured the Ge Pin voltage is high when high temperature Ge Pull up done.

    start fail reset ok.txt

  • 0 in reply to user5874521
    TI__Genius 10370 points

    I think you are miss understanding this feature. The bq35100 power does not flow though the senseR since the device ground is on the cell side. This means the gauge power is not measured. 

    To combat this the bq35100 adds the average gauge power to account for when the gauge is on and active. As Kang said though, the gauge has many different modes of operation so it does not account for all apects and it is only an average. You can remove this by writing the DF to whatever value you want or to 0. 

    Thanks,

    Eric Vos

  • 0 in reply to Vos
    Intellectual 450 points

    Hi Eric,

    Thank you for your support.

    "The bq35100 power does not flow though the senseR since the device ground is on the cell side." But in the spec and BQ35100_EVM borad, bq35100's ground is on the pack- not cell side(BAT-). Emmm, or the cell side you said is Pack-?

     Then in the page 9 BQ35100_TRM, it said, "The measured current also includes the current consumed by the device. "

    You said, To combat this the bq35100 adds the average gauge power to account for when the gauge is on and active.

    But TRM said "Current() = Actual Measured (SRP–SRN) Current – EOS Gauge Load Current" And default EOS Gauge load Current is a positive number +0.35mA, it is not adding but subtracting.

    TRM said the Actual Measured (SRP–SRN) Current is the whole consumption include the bq35100. And the Current() is the value showed to users.

    It is not easy to understand.

    In addition,

    How to write this DF, I can't find this register anywhere, what is it's address? If I haven't written anything on it. Is the default value 0.35mA? Will bq35100 add 0.35mA automatically？Or it will not add this default value? 

  • 0 in reply to Vos
    Intellectual 450 points

    Hi Eric,

    Thank you for your support.

    BQ35100 don't measure the current consumed of itself. Due to I can't find the address of the DF. I want to know if the value in DF has been added in the current or not. If I didn't write anything on this DF, will bq35100 add it automatically? Is the default value in this DF 0.35mA or 0 or other value? 

  • 0 in reply to user5874521
    TI__Guru 52115 points

    I will close this thread since I think we already provided our answer via email.

    Andy

  • 0 in reply to Andy Liu (BMS)
    Intellectual 450 points

    Hi Andy,

    Sorry, We still have trouble to use Advanced Comm not only in our own product but also in BQ35100_EVM (demo board).

    We have to use Advanced Comm in mass production. Please help!

  • 0 in reply to Andy Liu (BMS)
    Intellectual 450 points

    Hi Andy,

    I am sorry to bother you so much.

    We have trouble in Using I2C Communications With BQ35100.

    We consulted the steps in slua790 Page 8,9 and 10, but we failed to use Data Flash Access Commands to Change the Operation Config A. 

    We failed even on the BQ35100_EVM board. We have to use data Flash Access Commands to do calibration and test in mass production.

    Could you have a try on your BQ35100_EVM boards?

    Maybe the spec has left out some steps?

  • 0 in reply to user5874521
    TI__Genius 10370 points

    From the image your device is sealed. Calibration will fail when sealed because data flash cannot be updated. Once a device has been sealed once every time it is powered down or reset it will boot in a sealed state again. My suggestion would be to unseal, full access, then reprogram the bq35100 with the default firmware file found on TI.com (v1.02) 

    Once programmed fresh the first thing i would do is calibrate since the gauge will be in the right mode (unless you need to switch to multicell config). After i would finish the other configuration options you are needing to change. 

    Thanks,

    Eric Vos