• State
  • Locked Locked
  • Replies 15 replies
  • Subscribers 64 subscribers
  • Views 1601 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.

OPAMP feed to bq34z100G1

Peter Upczak
Intellectual 965 points
Other Parts Discussed in Thread: INA326, INA188, INA333, BQ34Z100-G1

Hi Tom -- I indicated that we have some success with a breadboard interface using an an opamp to boost our 0.5mohm sense resistor to the levels you suggested we'd get by using a 1.5mohm sense resistor (back in this thread: )

All seems to be going well with the active opamp interface, I'm still doing optimizations to see how "good" we can get even with this sloppy hand-tacked breadboard using external dc power.

Looking forward, we'd like to replace our active A/D feed from that 0.5mohm resistor, a Burr Brown / T.I. INA326 instrument amp, with something more to date. The T.I. INA 188 looks like a drop-in replacement requiring only 2 or 3 passive component changes to meet our needs. Assuming all power supply / conditioning, decoupling, and dc offset and voltage clamping to the BQ is handled, do you think the INA 188 is a good choice for this? There are many choice in the INA family, this one just looks great for our purpose.

If not, would you be able to suggest another suitable amp?

thanks, -pete u.

  • 0
    TI__Mastermind 27095 points
    Hi Peter,

    INA188 looks to be a suitable replacement for INA326 in your application. I forwarded your question to the instrumentation amplifier group as well. I will let you know if they have any comments.

    Best Regards,
    Ben Lough
  • 0 in reply to Benjamin Lough
    TI__Mastermind 27095 points
    Hi Peter,

    Take a look at INA333. It has a much lower input offset voltage than INA188 and is more recent than INA188.

    Best Regards,
    Ben Lough
  • 0 in reply to Benjamin Lough
    Intellectual 965 points

    Benjamin, Thank you for that information. I'll have a closer look at the INA 333.  -pete

  • 0 in reply to Peter Upczak
    Intellectual 965 points

    Hello T.I. team:

    For the bq34z100g1 Pin Function descriptions, the datasheet states

    SRP:

    Analog input pin connected to the internal coulomb-counter peripheral for integrating a small
    voltage between SRP and SRN where SRP is nearest the BAT– connection

    SRN

    Analog input pin connected to the internal coulomb-counter peripheral for integrating a small
    voltage between SRP and SRN where SRN is nearest the PACK– connection

    The attached HIGH-LEVEL CONCEPT schematic, quadrant B2 depicts the current-sense approach I'm given to work with. Note the placement of AGND reference and Pack - connection. I have to stick to that arrangement due to requirements owing to the larger system that provides and consumes battery charge.

    Although the arrangement in B2-B1 seems" to work in concept/bench testing, does this violate bq34z100g1 design mandates for SRP and SRN affinities?

    Alternatively, quadrant A2 shows how I might invert this to satisfy at least the SRP connection, but it leaves SRN still questionable per the Pin Function description AND imposes the second amp which I'd rather avoid. (currently looking at INA 188 and 133).

    Thoughts, comments, recommendations appreciated. Thanks!

    BQ_Concept.pdf

  • 0 in reply to Peter Upczak
    TI__Mastermind 27095 points
    Hi Peter,

    I have not seen the bq34z100-G1 coulomb counter used on the high side before. Let me talk with some of the designers to see if they have any concerns or input to provide on your implementation.

    Best Regards,
    Ben Lough
  • 0 in reply to Benjamin Lough
    Intellectual 965 points
    Thanks Ben. I'll sit tight for now.
  • 0 in reply to Peter Upczak
    TI__Mastermind 34775 points

    Hi Peter,

    Are U1 and U2 INAs? Interestingly, your drawing depicts them as OPAMPs with positive feedback. I think you need to clean this up. Also U3 I'm guessing is a buffer, but the feedback is connected to the non-inverting (+) input of the OPAMP, but should be connected to the inverting (-) input for a buffer.

    What voltage or voltages you are using or plan to use for the INA and OPAMP supplies? Dual or single supply? It would have to be a dual supply (+ and -) with common mode ~0V to work with the bq34z100-G1 SRP and SRN input voltage spec.

    You said you had this working on a breadboard setup? Can you provide details/schematic of the breadboard circuit used? 

  • 0 in reply to Damian Lewis
    Intellectual 965 points
    Hi Damian -- yes my drawing is a very loose depiction of circuit elements, intentionally. I will put together an exact schematic of the interface arrangement I'm experimenting with and send it along. There is an INA326EA in front of our sense resistor, on a +/- 2.5v supply to our board ground. INA 188 or 333 is under consideration for nextgen, but I've not gotten that far yet.
    I'll try to have my details to you tomorrow.
    Thanks. -pete
  • 0 in reply to Peter Upczak
    Intellectual 965 points

    Damian -- I hope this clarifies things for you. thanks! -pete

    BQ34z100g1_EVM_interface.zip

  • 0 in reply to Peter Upczak
    TI__Mastermind 34775 points

    Hi Peter,

    Thanks for the files. I will review them next week. Can you share with me the reason behind this active/passive approach?

  • 0 in reply to Damian Lewis
    Intellectual 965 points
    Hi Damian - our existing design produces just under 2W across 0.5mohm Rsense at 60A which in typical application is sustained for 10 to 12 minutes. The battery pack & control board are tightly packed and fully sealed, so things get hot. Our design margin is 90A, we're fused at 80A, short bursts are not uncommon. Using a 1.5mohm in the same space is thermally prohibitive.

    Normal operation also includes being down in the range of 5A or less. I believe the 1.5mohm Rsense target was recommended to get the low-end of the input scale out of the BQ noise plane, while capping at around the 0.125V recommended operating input voltage of the BQ at our high-end peaks. (btw this is where I started reading too much into the relationship between BQ Vin's and the 32767 limit).
    Since we already have active translation, I simply decided to follow that as a way to satisfy the parameters at both ends. It would certainly be much easier to use a real 1.5mohm Rsense.

    regards, -pete
  • 0 in reply to Damian Lewis
    Intellectual 965 points
    Hi Damian - our existing design produces just under 2W across 0.5mohm Rsense at 60A which in typical application is sustained for 10 to 12 minutes. The battery pack & control board are tightly packed and fully sealed, so things get hot. Our design margin is 90A, we're fused at 80A, short bursts are not uncommon. Using a 1.5mohm in the same space is thermally prohibitive.

    Normal operation also includes being down in the range of 5A or less. I believe the 1.5mohm Rsense target was recommended to get the low-end of the input scale out of the BQ noise plane, while capping at around the 0.125V recommended operating input voltage of the BQ at our high-end peaks. (btw this is where I started reading too much into the relationship between BQ Vin's and the 32767 limit).
    Since we already have active translation, I simply decided to follow that as a way to satisfy the parameters at both ends. It would certainly be much easier to use a real 1.5mohm Rsense.

    regards, -pete
  • 0 in reply to Peter Upczak
    TI__Mastermind 34775 points
    Got it Pete...Thanks for the explanation!
  • 0 in reply to Damian Lewis
    Intellectual 965 points
    Hi Damian -- checking in to see if you've made any progress on the high-side/low-side GND thing ...
    Also, I was just reading SLUSBZ5B page 35 (sec 8.3.7.1) description of GNDSEL bit in Pack Configuration Register, I wasn't completely aware that I could select SRN as my ground reference. Does that imply that SRP could be used as the "active" level for monitoring current, opposite of the reference designs? I'm not sure what I'd do with that, it's a curiosity for me at this point.
    regards, -pete
  • 0 in reply to Peter Upczak
    TI__Mastermind 34775 points

    Hi Pete,

    This bit applies to the ADC GND reference to select between VSS and SRN. Typically VSS and SRP are the same (BAT-). By setting GNDSEL to 1 selects SRN as the ADC GND, which can flip the polarity of the current measurements and add the voltage drop across RSENS to the VBAT measurements.