• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 63 subscribers
  • Views 895 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.

BQ51050B: RX does not start to charge

Wout Geeurickx
Prodigy 235 points
Part Number: BQ51050B

Hello,

I am currently trying out the bq51050BEVM-764 for a project where there are heavy size constraints. We previously used another Qi RX chip from another manufacturer but this chip was way too complex for our purpose and it took too much useful space on the board. Because of the size constraints, we are using a WR121210-27M8-ID coil from TDK. The main reason behind this design choice is because it is very thin. Some of the parameters of the coil worry me however: the inductance is rather low for the diameter of the coil and the resistance is also quite high. This even causes Q to be around 55, lower than the required 77 from the Qi standard. However, the manufacturer of the coil ensured us it would work for Qi charging and this was already validated using the previous Qi RX chip. 

I removed the original coil from the EVM, soldered my new coil to it and replaced the tuning caps. Tuning caps are chosen to be Cs = 288nF and Cd = 2.8nF, based on simulations performed in LTSpice. I also changed the COM caps to 68nF as was suggested is some other threads.

As the figure below from a measurement on VRECT suggests, there is some communication but then the voltage drops again. This figure is from a measurement without a battery connected, another measurement with battery showed the same curve, only the voltage dropped earlier: around the moment that the curve from this plot becomes flat before the two last spikes.

 

Is there an explanation for the almost triangluar waveform before and after the increase in VRECT?

Could the parameters of the coil be the reason why the charger does not start charging the battery? As we only have a small battery, the Ibulk is set to 100mA with a termination current of 20%. Is this too low for accurate operation?

Best regards,

Wout

  • 0
    TI__Guru* 75656 points
    Hello
    The scope capture show a good start up (first three pulses) and tuning (pulses every 30mS) to desired operating point, then load in enabled.
    A pulse then a drop in output suggest the RX sent an End Power Transfer (EPT) to turn off the TX. Also the TX does not try to restart the RX, low power ping checking if the RX has been moved (Triangluar waveform).
    Any LED on the TX indicating fault or charge complete?
    Which TX are you using?
  • 0 in reply to Bill Johns
    Prodigy 235 points

    Hello Bill,

    Thank you for your fast response! The TX is a custom-made design of which I do not know much details. It was verified that it did work before (with the other Qi RX chip)There is a feedback LED present which is blinking during charging (but also when the battery is full). I see this LED blinking about 2 times and then the LED goes back to constant (waiting for RX). I have added the scope capture of both the VRECT (TP12, top) and CHG (TP10, bottom). The LED on the EVM also blinks twice (as indicated by the two low values on the graph).

    I also did a scope measurement with another off-the-shelf RX (Qi charger for Moto 360 Smartwatch) and this came out. Another response, but still no battery charging.

    Does this clear things up for you?

    Best regards,

    Wout

  • 0 in reply to Wout Geeurickx
    TI__Guru* 75656 points

    Hello
    Do you have a TI EVM on hand?
    I can send some software that can report what the TX is doing.

    Both TX seem to try and start charging but then TX will stop, possibly on command from the RX.
    What does the battery voltage look like?

    Can you send RX schematic?

  • 0 in reply to Bill Johns
    Prodigy 235 points

    Hi Bill,

    Your comment on the communication and tuning looking good, made me think about current settings. I am just using the EVM with adaptation to the coil, tuning caps and COM caps. I was checking all the jumpers and potentiometers and then I came to the R13 pot for the termination current. The battery I used was at 3.8V, which I thought was low enough but apparently it isn't. The current must have fallen immediatly below the termination current. When I lowered the R13 pot for lower termination current, the battery seemed to charge again.

    Thank you for bearing with me and verifying the VRECT measurements.

    This solution brings up another concern. As I mentioned in my initial post, we are using the charger for low current charging. The battery we are using has 60mAh capacity, so normal 1C charging would mean 60mA charging. I read on this forum that the IC might have some problems regarding stability when used for this low charging currents. Is there a work-around possible to enhance stability? For this particular application, charging is not time-critical and can sometimes last a full day. It may be an option that we short the TERM pin, but in this case I was wondering what happens when the 10 hour timer runs out? Does this also issue an EPT?

    Best regards,

    Wout