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: Output and regulation messed up (with TPS63031)

Part Number: BQ25505
Other Parts Discussed in Thread: BQ25504
This is our power path circuit. We use a single solar sell and 3.7Vdc lithium battery. 
First thing we experienced was that sometimes the power regulator failed to start and it's output (VOUT) was just a few mV. This was solved by removing the 10uF capacitor (C10) at TPS input (VIN_B).
Then we decided to take off the switch and replace it with a jumper (0 ohm resistor between VIN_A and VIN_B), so the circuit stays "always on".
Randomly, 3 things happens (always with battery connected): 
1. No light input on solar cell (photocell_Vin = 0) -> harvest output (VIN_A) 0.5 Vdc;  Regulator output 1.6 Vdc;
2. No light input on solar cell -> harvest output 0.004 Vdc; Regulator ouput 0.3 Vdc;
3. Light on solar cell (about 3.2Vdc photocell_Vin) -> harvest output 1.5 Vdc, Regulator output 0.5 Vdc.

Just once it worked fine (harvest output 3.7Vdc; Regulator output 3.3Vdc).
Case 1 is the most common. Case 3 is the most bizarre for me. 
I've tried to remove all capacitors one by one (C5, C7 and C32), didn't work either.

Can anyone help?
  • Caio,

    Does the battery have a protection circuit? What is the current coming from the solar panel? Is there a DC supply you can use to try instead of the panel to test the circuit?

    Also is the 3.2V the OC voltage of the panel or is it when the load is present?

    regards,
    gautham
  • I'll verify those questions today and get back with answers asap.

    Thanks,

    Caio

  • 1. I believe the battery does have a protection circuit.
    2. On the measurements without light there is no current. We did not measure the current coming from the solar panel. If really necessary, I could try to do so.
    3. Unfotunately not right now, but I'll get my hands on one soon (ours just stopped working a couple days ago).
    4. 3.2 V when connected to the harvester while under bright artificial light. Under the same light we got 4.7 V open circuit.
    The point is: it has been working fine with the mechanical switch, which just physically connects harvest output to the voltage regulator input (if you consider the voltage reg, mcu and sensor as a block then this is the load form the harvester point of view). Why does jumping the switch would change it's behavior?
  • Caio,

    We will need to make sure the harvester is able to provide beyond the Iq consumed by the TPS. it has around 50uA of quiescent current. Also I am not sure the behavior of the TPS under the UVLO voltage.

    Having a switch is truly isolating the output of the BQ and the input of the TPS. Is it possible for you to use the VBAT_OK signal to drive a load switch or the EN pin of the TPS?

    regards,
    gautham
  • Gautham said:

    We will need to make sure the harvester is able to provide beyond the Iq consumed by the TPS. it has around 50uA of quiescent current. Also I am not sure the behavior of the TPS under the UVLO voltage.

    The point is, there is a fully charged battery connected to the BQ. So, as I understand, the BQ was supposed to deliver, at least, the voltage and current available at the battery. 
    Do you suspect that, somehow, the BQ is failing to source current so the connection between the BQ output and TPS input is driven low?

    Why does it work sometimes (at least once)? Actually, the problem is not really about the TPS behavior under voltage, I really want to understand is why the BQ can't output the expected voltage with a fully charged battery connected.

    Gautham said:

    Having a switch is truly isolating the output of the BQ and the input of the TPS. Is it possible for you to use the VBAT_OK signal to drive a load switch or the EN pin of the TPS?

    We could take out the capacitor C11 (that is grounding both VINA and EN) and connect to VBAT_OK, but according to the TPS datasheet, I believe this capacitor is necessary for VINA.

    Thanks for the answers,

    Caio

  • Caio,

    I found that section 7.4.3 on the datasheet has a guideline on connecting a charged element. 

    It might be good to discharge VTSOR node to GND before reconnecting the battery as an experiment. 

    thanks,

    gautham

  • We managed to put a transistor between BQ and the load/TPS, driven by VBAT_OK, removing the mechanical switch. The schematic now is as follow:

    The system still behaves weirdly.

    With the solar panel obscured, we plug in the battery and the system started fine, but then we removed the battery and plugged it in again - no good. VBAT_OK was 0 and BQ output hovers around 450mV, slowly going down to around 65mV. If we somehow "discharge" the BQ output (either shorting it or waiting for it to slowly discharge) before plugging back in the battery, the systems works fine.

    What might be happening? Is this expected? If yes, what am I missing?

  • Caio,

    I am digging up the history on the datasheet comment. Meanwhile I think you are seeing the symptom reported on the datasheet.

    Are you able to see the system come back up if the solar panel was not obscured and the battery was plug/unplug?

    regards,
    gautham
  • So, if I understand correctly, this seems to be a somewhat expected behavior, right? If so, any ideia how to work with/around it?

    We are looking into the BQ25504, that seems to be simpler and more appropriate, since we only have one battery in the system, but maybe it has the same issue regarding plugging in the battery "in the dark".

    Anyway, I'll do the tests with the solar panel "on" and get back to you.

    Thanks,