• State
  • Locked Locked
  • Replies 29 replies
  • Subscribers 63 subscribers
  • Views 7007 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.

BQ24450: State Machine and Output Not in Sync

DavidC482
Intellectual 280 points
Part Number: BQ24450

Hello.    I've successfully built a charger around the BQ24450 and  completed some preliminary testing.  I've noticed that the State Machine output (S1 and S2) do not appear to be in sync with the output.  The output has clearly gone into Float Mode - the Vfloat is sitting around 13.96 (a bit high) and the current has significant dropped to <100ma (Float Mode, also a bit high) .  S1 and S2 are both low (<.15V) which is the state for boost mode.  So they are not in sync.

What's going on here?

2nd question - I'd like to be able to that the charger is in deep Recovery Mode (charging when battery <7.5V).  I was hoping to use the pre-charge output (pin11), but that doesn't seem to be working.  Even in float mode, that output is 14.6V.  Do I need to tie other pin conditions together to be able to show a Recovery Status.  If so, what might be recommended?

Thanks,
Dave

  • 0
    TI__Mastermind 21590 points
    Hi,
    I'm looking into this. I'll let you know when i have an answer.
    Regards,
    Raheem
  • 0 in reply to Raheem
    Intellectual 280 points
    Hi Raheem,
    Have you figured out why the state machine and the output are not in sync?

    And to the 2nd question, I'd like to be able to show that the charger is in deep Recovery Mode (charging when battery <7.5V). I was hoping to use the pre-charge output (pin11), but that doesn't seem to work, as the pre-charge output is high even in float mode. Is there a condition or set of pins I could use to show a battery is being recharged using pin 11 (Pre-charge)?

    Thanks,
    Dave
  • 0 in reply to DavidC482
    Intellectual 280 points
    Raheem,

    I exchanged the BQ24450 with a different one and got very silmilar results - the S1 and S2 output states are not in sync with the output. My output is clearly in float mode but, initially, S1=S2=0V (Boost Mode). And after charging overnight, S1(pin-10) was quasi-hi @2.51V and S2(pin-9) was still low @0V. This corresponds to Bulk Charge mode.

    Regardless of the mode, there is a systemic problem where the state machine does not reflect what's going on at the output.

    So do yo have a Batt Charger chip that is a good/better substitute for the BQ24450 that also shows charge state status? Because this one doesn't report correctly.

    David.
  • 0 in reply to DavidC482
    TI__Mastermind 21590 points
    David,
    Can you please post your schematic?
    Regards,
    Raheem
  • 0 in reply to Raheem
    Intellectual 280 points

    Raheem,

    Here ya go...All the i/f ICs are CMOS so there is no load on the charger chip.

    Dave


  • 0 in reply to DavidC482
    TI__Mastermind 21590 points
    Hi,
    Can you please post a clearer image. This one is hard to see.
    -Raheem
  • 0 in reply to Raheem
    Intellectual 280 points

    Raheem,

    To make it clearer (less blurry when zoomed in), I'll have have to capture it in parts.  Do you care about the interface CMOS chips?

    Dave

  • 0 in reply to DavidC482
    TI__Mastermind 21590 points
    David,
    I'm currently only trying to look at the charger IC.
    Regards,
    Raheem
  • 0 in reply to DavidC482
    Intellectual 280 points

    Raheem,

    Your media insertion tool reduces the resolution.  It starts out very clear when I capture the image, but by the time it gets posted here, it's not as clear.

    Let me know if this is still not clear enough.  If still not clear enough, I'll have to break and post this in 2 parts.

    Dave

  • 0 in reply to DavidC482
    Intellectual 280 points

    Hi Raheem,

    Now that you've had a chance to examine the schematic, have you been able to identify the issue?  Or are you as baffled by this as I am?

    Thanks,

    Dave

  • 0 in reply to DavidC482
    TI__Mastermind 21590 points
    David,
    Looking at your schematic, your charge current is around 9mA which is less than the minimum of 25mA. Please adjust R2 to fix this.
    Regards,
    Raheem
  • 0 in reply to Raheem
    Intellectual 280 points
    Hi Raheem,

    Thank you for looking into this. Not sure what R2 is, but assuming it's Rt, then this is interesting. A few observations:

    First, my take-away from your response is that the Ipre-chrg is going to have an effect on the STAT1/STAT2 state machine output. Is that correct?

    Some verification calc's:
    1. In the BQ24450 data sheet, dated Apr 2009, in the example at the end (this may be obsolete info), the example shows a an Ipre-chrg of just 10mA (below the 25mA min).
    2. Using your 9mA, my 62R, I solved for Vpre across Q5. [Vin=16V; Vbat(min)=5V], I get the voltage drop across Q5 when turned on to be 5.42V. This is way too high for an 'on' volt drop.
    3. Finally, assuming a voltage drop across Q5 to be .5V(worst case), solving for Rt [Ipre-chrg=50mA; Vin=16V; Vbat(min)=5V], I get an Rt of 300R. Is this what you are looking for? This implies raising the resistance will increase the current Ipre-chrg. Hmmm...

    Just trying to get on the same page with you.

    Thanks again for all your help.
    Dave
  • 0 in reply to DavidC482
    Intellectual 280 points

    Raheem,

    Ignore everything I said previously. R2 is Risns. How did you get 9mA at what mode (state)?

    I've got a current meter showing current into the battery. In float mode, I get exactly 50mA, the calculated value.

    What am I missing?

    Thanks,
    Dave

  • 0 in reply to DavidC482
    Intellectual 280 points

    Hi Raheem,

    Let me add to my previous statements.  My calculations for Imax input current is as follows:

    Imax = (Vin - Vbatt - Vdiode - Vqext)/R2

             = (16 - 10(assume) - 1 - 1.5(approx))/.27R

             = 3.5V/.27 ~ 13A!

    Maybe I don't know how to run the calc's, but I'm not getting anywhere near 9mA.  I was shooting for 1A.  Got close.   In actuality, I get the following current in each of the modes:

    Bulk = .9A  (nearly per the calc's)

    Boost = .9A

    Float = 50-100mA  (per the calc's)

    Pre-Chrg = ?    (Don't know.  Haven't drawn the battery low enough)


    What I have not been able to find is the value for Vilim.  the write-up is ambiguous.  Maybe that's my disconnect. Nonetheless, my .27R is in the same magnitude as that shown in the example problem ( .42R).  

    Thanks,
    Dave

  • 0 in reply to DavidC482
    TI__Mastermind 21590 points
    Dave,
    Imax-chg = 250mV / 27 ohms = 9.25mA . That's where i calculated your charge current to be 9mA. IPRE = (VIN – VPRE – VBAT) ÷ RT.
    Regards,
    Raheem
  • 0 in reply to Raheem
    Intellectual 280 points

    Hi Raheem,

    Thank you for getting back to me on this.  Now I see where you got 9mA.    It's hard to see, but Risns (R2) is really 0.27R, not 27R.   Given that the calc is off by 100, that accurately predicts the 900mA I am actually getting.

    Ok, so back to the problem at hand.  Why is the state machine not reporting the correct status of the battery charger?

    Thanks,
    Dave

  • 0 in reply to DavidC482
    TI__Mastermind 21590 points
    David,
    Can you post scope shots of VBAT and both STAT pins when this issue occurs?
    Regards,
    Raheem
  • 0 in reply to Raheem
    Intellectual 280 points

    Hi Raheem,

    Yes, but it'll take a little bit of time..  Step 1 is to drain the battery.  Is there a voltage you want me to start at?

    Thanks,
    David

  • 0 in reply to DavidC482
    TI__Mastermind 21590 points
    David,
    If it's possible, it'll be nice to see the whole charge cycle of see what the stat pins show.
    Regards,
    Raheem
  • 0 in reply to Raheem
    Intellectual 280 points

    Hi Raheem,

    I'm working on it.  I'm draining the battery. Started about 5 hrs ago.   I'm in the heart of the battery charge @12.48V. At the current rate, it could be most of the weekend to get it down to 10.5V.  That's low enough for bulk charge status, but not recovery charge.  Is that good enough? Tomorrow morning, I'll increase the load to make it drop faster.

    Thanks,
    Dave

  • 0 in reply to DavidC482
    Intellectual 280 points

    Hi Raheem,

    Iver the weekend, I depleted the battery and performed a recharge process.  I have pic's and data in raw form. 

    In a nutshell, everything worked as it was supposed to in bulk and boost mode, but in float mode, the Stat1 (pin10) does not come up to 5 V.  When Stat1 goes high, it's at  a 1.5V level, so my logic still sees it as 0V, and so instead of the state machine going to a 1/1 state(float), it looks like it goes to a 0/1, which is bulk charge mode.   Is this 1.5V normal?  I wouldn't think so, but if it is then I have to use this 1.5V as high and not 5V.

    How would you like this data/pics presented.  Word, Excel, something else?

    Thanks,
    Dave

  • 0 in reply to DavidC482
    TI__Mastermind 21590 points
    Dave,
    Pictures should be good. Also what do you have STAT2 pulled up to? It's not shown where the wire goes in your schematic.
    Regards,
    Raheem
  • 0 in reply to Raheem
    Intellectual 280 points
    Hi Raheem,
    I'll get those pic's prep'd and posted shortly. I need to show where the Gnd line is.

    As for Stat2, it's pulled up to Vcc(+5V) thru a 5k resistor. That's how I'm able to get the 5V signal.

    It's Stat1 (pin 10) that's the problem. The state machine is only putting out 1.5V.

    I'll get the pic's ready.

    Thanks,
    Dave
  • 0 in reply to DavidC482
    TI__Mastermind 21590 points
    Dave,
    What is the voltage on Vfb when this happens?
    Regards,
    Raheem
  • 0 in reply to Raheem
    Intellectual 280 points
    Raheem,

    To the Vfb question when the charger goes into Float Mode [Stat1= 1.5V; Stat2=5V], Vfb=2.37(init)->2.34(when probe pulled off). So if it's supposed to be much closer to 5V, it's not.

    Had to go back and retake a better Bulk Chrg pic and wait for the Float Mode to occur. Will post them shortly.

    Thanks,
    Dave
  • 0 in reply to DavidC482
    Intellectual 280 points

    Hi Raheem,

    Still working on the pics.  Also, pulling together a parts list for the batt charger.

    Can you tell me if a power Schottky is allowable? .  It generally has a lower Vforward voltage drop.

    Thanks,
    Dave

  • 0 in reply to DavidC482
    TI__Mastermind 21590 points
    David,
    The reason i'm asking is that in float mode, both stat1 and stat2 should be Hi-Z mode. Stat2 makes sense to be at 5V because the Q8 BJT is off and it's pulled up to the 5V rail. Stat1 should be at around 2.3V becuase the Q9 BJT is also off and that pin is pulled up to Vfb which means you should be around 2.3V on that pin in float mode.
    Regards,
    Raheem
  • 0 in reply to Raheem
    Intellectual 280 points

    Hi Raheem,

    Now we know where the issue lies.  If Stat1 gets pulled to Vfb(2.3), then Stat1 will never be a true "high" level.  Stat1 high is a 1.5V signal when high.  If this was stated in the data sheet that Stat1 hi was 1.5V, then I missed it.  So now I need to make an interface that translates 1.5V to 5V.  That will give me the status I am looking for.  I can use either a diff amp or an P-Ch FET switch to 5V to create that high signal level that would feed my CMOS logic.

    I'll keep you posted on how it turns out.

    Thanks,
    Dave

  • 0 in reply to DavidC482
    Intellectual 280 points
    Raheem,

    A question I had asked once before, what can I use to detect the deep recovery recharge mode? This is Vbat < Vth level. As long as Vbat is < Vth, then I want to indicate a deep recover mode. What combination of outputs or pins can I use?

    Thanks,
    Dave