• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 463 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.

BQ25505: Capacitor discharge rate or discharging current with no load. power for cold start

Juan Romero
Prodigy 210 points
Part Number: BQ25505

This question refers to the BQ25505EVM. I have been trying to use this board to harvest power from a 50uW, 100kOhm source. However, measurements seem to indicate that the board requires much more current and power than advertised (~3.5mA and 1.4mW) to get to Vstor=1.8V. For that reason, I have been looking at the discharge rate of the on-board capacitors with no load (only the 1Meg oscilloscope to measure the voltage). I noticed that the capacitor discharges very quickly, VSTOR goes from 4V to 2V in 45s which translates to a peak discharge current (calculated) of 10.9uA. . Do you have any comment on this? May I be making wrong assumptions or a better setup is advisable for this task? At this point, I am looking for an explanation for the excessively high current requirement of the IC although I am aware it is very unlikely I could use it for my application.

  • 0
    TI__Guru**** 165766 points

    HI Juan,

    Unfortunately, we do not have the cold start minimum power fully characterized.  The minimum input voltage from the source needs to be 600mV.  Regarding the discharge current in boost charge mode, I can't explain the discharge.  The charger is powering from VSTOR with Iq~=350nA but periodically attaches the 10Mohm feedback resistors at VRDIV.  Did the PFET between VSTOR and VBAT turn on so that the 100uF discharges to VSTOR or is the PFET turned off so that only the 4.7uF capacitors on VSTOR is discharging?

    Regards,

    Jeff

  • 0 in reply to Jeff F
    Prodigy 210 points

    Yes, the 100uF discharges to VSTOR.

  • 0 in reply to Juan Romero
    TI__Guru**** 165766 points

    Juan,

    When I an EVM on the bench with VIN_DC floating and a 4.0V Keithley source meter on VBAT=VSTOR, I only see about 450nA average leakage current from the source meter into the device.  So, with I=c*dv/dt, the 100uF capacitor should stay up for 444 minutes.  When I remove the source meter and measure the VSTOR and VBAT on the scope with scope probes having 10Mohm impedance on VSTOR and VBAT, I add 4V/10megohm = 0.4uA*2=0.8uA more leakage at the start but that decreases.  Mathematically I expected 160s but I saw 200s from 4V to 2V due to the decreasing leakage through the scope probe and also into the device.  Note:  I removed the 100uF ceramic capacitor on BAT_SEC in case it had higher leakage current but it didn't make much difference.

    Regarding cold start, I also tested to see if the typical cold start power of 15uW was reasonable.  So, with my Keithley set to 600mV, I increased the current compliance until VSTOR would reach 1.9V with no other load on VSTOR and only the 100uF on VBAT.  The sourcemeter needed about 22uA, which would be 600mV*22uA = 13.2uW.

    Regards,

    Jeff

  • 0 in reply to Jeff F
    Prodigy 210 points

    Hi Jeff,

    Thanks for taking the time to this, I really appreciate it. I can agree with the first set of measurements. Your setup seems more accurate than mine and I did measured around 450 nA from the SMU using this approach..

    Regarding cold start, my Keithley reached compliance at 3mA and still VSTOR was only 1.56V. I only waited a few seconds for the capacitor to charge but I think that was enough. Also, I measured VSTOR using scope probes of 1Mohm but the effect should be minimum. I have to mention that the measured voltage was only 0.38V (even though it was set to 0.6V) but that is because the EVM is drawing so much current that reaches compliance, thus limiting the voltage (does this makes sense?). The jumpers of the EVM are set to /EN-GND and VOC_SAMP-80%. Do you think the problem could be at the EVM?

  • 0 in reply to Juan Romero
    TI__Guru**** 165766 points

    Juan,

    The cold start circuit actually plans the input voltage when it is running so seeing on 0.38V with 0.6V applied is expected.  In cold start, the jumpers do not matter.  Cold start charge pump is very inefficient.  It can take several minutes for the capacitor to charge.  It took mine about 5 minutes.

    Regards,

    Jeff

  • 0 in reply to Jeff F
    Prodigy 210 points

    Thanks again Jeff. That seems reasonable but even after >30min (and ~24h charging) my board won't start with 25 uA. At this point I have gave up on using the BQ25505.