• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 63 subscribers
  • Views 963 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.

BQ24650 HIDRV, PH nodes strange ringing -> device not working

Juan Medrano83
Intellectual 535 points
Other Parts Discussed in Thread: BQ24650

Hello,


I am solving several field failures of a device using the BQ24650 that is already in production. Some of them have rather strange behaviour in the HIDRV and PH nodes. In this group of failures, the devices have all signals working correctly, with exception of HIDRV, PH  and MPPSET . Please see the following scope captures:

HIDRV pin reference to GND

Vph referenced to ground:

So this waveforms are not of normal operation of the device. I understand HIDRV is 6V higher than PH when enabling the High side mosfet. So here is a waveform of HIDRV referenced from PH, which is exactly Vgs of the high side mosfet, for reference:


What could be the cause of that strange waveform that 1) has that ringing and 2) is driving the high side mosfet for a very short time. I think this is the reason why MPPSET can not be correctly performed and is at 1.4V instead of 1.2V.

Following are waveforms of all other related signals for you to understand the status of the device:

Vts referenced to GND  (seems OK)

**************************************************************************************

Vfb referenced to GND (sems OK)

**************************************************************************************

Vref referenced to GND ( ~3.3V, seems OK)

**************************************************************************************

Vregn referenced to GND( ~6V, seems OK)

**************************************************************************************

Vmppset referenced to GND (1.32V, sometimes 1.4V, not ok)

**************************************************************************************

Vbtst referenced to Vph, (not OK, should be very stable at ~6V)

Vbtst referenced to Vph from a known good board (

**************************************************************************************

Vcc referenced to Vsrn: Vcc >> ( Vsrn + 100mV) , actually Vcc is aprox 7.44V above Vsrn so the device is definitely ON.

**************************************************************************************

Vlodriv (does not look ok, just as hidrv, does not seem it is switching fast enough)


Input:

Solar Panel, 15W, Voc23V, Vmmpt ~ 16.5V, 10m cable length. I am using a real panel, not using a bench supply to simulate it. In either case the behaviour is the same.

Output:

Sealed Lead Acid battery, 12.6V, 12Ah,

Sense Resistor 20mOhm,

Per the charging enable/disable requirements listed in Page 14 of the BQ24650 datasheet, all charging conditions are OK.

STAT1 pin is LOW, indicating a charging is in process.

STAT2 Pin is high (14.34V).  The combination of STAT1 and STAT1 indicates NO FAULT DETECTED.
I can provide full schematic via private message. 

I will appreciate your help to identify the issue in this set of boards.

Thanks TI Team

  • 0
    TI__Genius 11480 points
    Hi Juan,

    Can you elaborate on what you mean by failure?


    From the waveforms, it looks like the converter is operating in DCM. Usually this is okay for a buck converter I think, but it may be causing you some grief here. BTST does seem a little "funky", though.

    You say you are charging a lead acid battery. Have you read this app note? It may be useful: www.ti.com/.../slva437a.pdf

    It is okay that MPPSET is above 1.2V. If you look at the system diagram in the datasheet, you will see that the 1.2V reference is at the non-inverting terminal of the op-amp. This means that MPPSET can be above 1.2V, but if it tries to drop below it will maintain at 1.2V. Comparing this to the Vfb op-amp, you see that the Vfb amp reference voltage is at the inverting terminal. That is why the output voltage can be below Vreg in the CC phase, but will not go above Vreg.

    If you'd like, you can send your schematic to d-wiest@ti.com. I will try to help you find any potential issues.
  • 0 in reply to David Wiest Jr
    Intellectual 535 points
    Hi David,

    Sorry, I lost my point on trying to make it clear. The problem is it does not charge the battery, this is no current flows from the solar panel to the battery in any situation. And for reference, for other known good working boards, the IC actively regulates input voltage to Vmp, which implies to set Vmppset voltage to 1.2V.

    I have already implemented the suggestions in the app note you mentioned. Thanks for pointing at it.

    So that is the issue. I sent you a friend invite to share documents with you.

    Thanks,
  • 0 in reply to Juan Medrano83
    Intellectual 535 points
    I do have seen Vmpp not at 1.2V, but not when Vbat < Vreg and Ipanel < Imax, this is, when the panel cannot provide as much energy as the charge controller is limited to, so that it will (or should) try to regulate input voltage to Vmppset to extract maximum power.