• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 64 subscribers
  • Views 477 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.

OK - Whence does the bq2060 get its power - And what is the true V(REG)?

Michael Banak
Intellectual 915 points
Other Parts Discussed in Thread: BQ2060, BQ2040, BQ2050

THe closer I look at this legacy chip, the more impressed I am. I can guess the wanswer to my question, but regardless of the reply, it looks like the bq2060 was a pioneering device in the area of flea-power design.

I see a REG output, which drives an N-Channel Depletion FET. That FET, in turn, powers even more circuitry, through an NPN Emitter-Follower. I can see that.

Here's my questions.

1) The internal REG circuit must get its power from *somewhere*. But Where is that source? Is it the RBI pin, with its diode and 10MEG Ohm in series? Is it one of the other Cell Inputs? My lands that internal regulator must be running off the earth's magnetic field.

2) The data sheet lists the REG voltage as 3.1V to 3.5V. Yet, I have reason to believe that this is actually the resultant Vcc voltage, appearing on the Source of the N-Channel Depletion FET In other words, that should be a listing for the resulatnt Vcc, especially since there is so much data about the JFET provided as a condition for assessing REG.

Please tell me the facts about these 2 issues.

Background:

I am studying a dreadful phenomena. I have a board exhibiting substantial swings in reported Load Current, even though no current is being pulled from the pack at all. This current (in the -10 to -15mA range, is highly dependent on both temperature and Pack voltage. The reported current is there, unchanged, even with Rsense shorted out.. I am therefore assessing everything connected to the regulators, etc., to account for everything I can in this system. I want to understand the function of the REG pin better, as so very much is derived from it.

I don't suppose this should matter, but I am seeing the bq2060_E207 in this board.

Yes, I have done VFC offset self-calibration three times, two of them under tightly controlled conditions.

Thanks,

Michael A. Banak

 

 

  • 0
    TI__Expert 6470 points

    Michael,

    The following is an excerpt from a recent email generated by one of my colleagues. The spreadsheet he refers to is attached.

    "All the older gauges that have a REF pin, show a very inexpensive NFET regulator in the typical schematics.  This is just a voltage follower and not a closed-loop regulator, so variations in the NFET as well as the very sloppy low-current zener that is bonded out to the REF pin, make for a very sloppy regulator that varies a lot with battery voltage and temperature.  Attached is a spreadsheet that contains the various tolerances and tempcos for the gauge.  If you plug in the min and max operating temperature, min and max Vgs(off) characterics for the NFET, and min and max battery voltage, the spreadsheet will yield the min and max Vcc range for a given bias resistor value."

    vcccalc.xls
  • 0 in reply to Doug Williams4811
    Intellectual 915 points

    Thank you, Doug. Your kind response has prompted 2 more questions. Sorry.

    - I still desire to know whence comes the power for this sloppy internal regulator. Into what pin is eneregy injected?

    - The Spreadsheet has an interesting warning:

    "Min bias current should be at least 2uA or min Vcc calculation will be less than calculation."

    If this "Bias Current" is delivered through the 10Meg REsistor, routing to the RBI pin, then the reference design (SLUU063a) is starving that regultor of necessaey bias current. What is this "Bias Current" of note? How does one juice it to insure 2uA? I think this plays into my first question.

    Gratefully,

    Michael A. Banak

     

  • 0 in reply to Michael Banak
    TI__Intellectual 940 points

    Michael,

    Some confusion here.

    The Vcc spreadsheet you were sent would be good for a bq2040, bq2092, bq20xx, or bq20xxH gauge that uses a REF pin.  The REF pin is essentially a low-current zener and the inexpensive reference regulator uses an NFET as a voltage follower.  There is no feedback in this circuit.  The following FAQ reference is applicable:  How to use the gas gauge REF pin to provide an inexpensive Vcc source

    The bq2060 does have a closed loop regulator built around an external JFET.  It does not have a REF pin, but does have a REG pin that can drive the gate of the regulator JFET without any need for an external bias resistor.  The on-resistance of the JFET is fairly high and can't supply enough current for LEDs, so circuits using LEDs are powered from an emitter follow connected to the JFET-regulated Vcc.  The following FAQ reference is applicable:  What is the criteria for choosing the proper Vcc regulator JFET?

    Bill

  • 0 in reply to Bill J10799
    Intellectual 915 points

    Thank you, Bill. I actually figured out the whole matter last nite, before taking a day off. I concluded the spreadsheet was for a bq2013 (or similar) and bq2050(or similar). The JFET self-bias circuit in the bq2060 is one that I use all the time as a chip designer (my "other life"). Once I figured it out, I felt kind of dense for not having thought of it.

    I have a problem to be addressed in another post up some day soon. After all calibration is done, I still get 10-15mA of current flow readings, even though there is no load.

    Once I finish my assessment of Vcc performance, I can frame those questions properly. Until then, thank you for clarifying and confirming the Vcc operation for me.

    We can count this thread closed.

    Best regards always,

    Michael A. Banak