• State Resolved
  • Locked Locked
  • Replies 14 replies
  • Subscribers 65 subscribers
  • Views 990 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.

BQ51003: BQ51003 Not Turning ON

Roberto Rodriguez
Intellectual 800 points
Part Number: BQ51003

Dear Support,

I am using the BQ51003 Qi WCP 1.2 compliant charger.  Values on the schematic are not the ones being used now that was based on the EVM board for the BQ51003. The device is enclosed in a 1mm thick PLA 3D printed enclosure with a coil WR222230-26M8-G 27.20uH facing the plastic, ferrite touching a battery, and the PCB on top of it.

I followed the steps to calculate the C1, C2 matching capacitors to tune the coil.

1. Measured the coil outside the enclosure free space = 27.20uH

2. placed the coil in the enclosure with all the components, battery and board.  Measured 25.92uH

3. Used the equations provided by the datasheet, and computed C1 to be 97.72nF - used 100nF.  C2 calculated as 940pF, used 820pF||120pF = 940pF

4. Soldered the components, 100nF rated at 16V (don't have 50V), others are 50V.  Not sure if this is the issue or not

5.  I place the device on the mat, and the mat starts blinking and blinking.  I put the scope at the RECT output, and I see it going up and I see also modulation like.  the LED connected at the output OUT blinks but won't go solid like when I put my cell phone on it.

6. Tried moving resistor ROS1(R37) to ROS2 (R36) 20K does not make a difference

7. Adjusted R34 392 ohms.  Originally 110ohms as the EVM - Didn't make a difference.

So what other troubleshooting steps can I follow?  Is the issue the coil not being tuned? The equations were followed to the letter - So why doesn't it lock?  If you have a recommendation of scope shot I can add that too.  I should also mention that after a while the LED stopped blinking and I could not get it do it anymore.  Not sure if the battery got to a full charge state and there is no more load current being pulled from the charger.   I will try with a discharged battery as well.

Any help will be greatly appreciated.

06/07/2021

Added a resistive load to the OUT pin at the decoupling capacitors.  The output ON TIME is 94ms every 2.5 seconds. it will not lock on always ON.  The Status LED blinks as following the pulses but it will not stay ON.  Any ideas what is happening?  is it communications, tuning, or otherwise?  Is there anyone that could help me resolve this?

VRECT  Magenta trace at the output decoupling caps of VRECT. the voltage swing is high.

6/7/2021 I can't reply to the threads for some reason.  

-I have seen and read the posts you sent thank you. 

- I don't know why it will not lock on.  The charger pad LED keeps blinking, the Status LED on the BQ51003 OUT pin connected to CHG_N does blink as well, and eventually stops.

-I have removed the thermistor and shunted a 10K in the TS-CTRL pin, and it didn't make a difference nothing... 

- I believe the coil is tuned per equations on the Datasheet.  I am not entirely sure why it refuses to cooperate.  I am using the correct coil 27uH.

- VRECT tries to go high and then shuts off.  I am not sure if too much power is reaching the Junction temp in the die and shuts off.  The load is pulling 200mA from it @ 5V.  It won't stay ON. 

6/7/2021 1:20PM PST

Instrumented better the VRECT capture .  Voltage appears to reach 10.10V approximately  See scope shot of VRECT.  VRECT magnitude ~10V, 70msec duration, and repeats around 800msec...

Read a different datasheet from another vendor where they explicitly say to place the assembly (Receiver) on top of the TX coil and measure Ls'  Equations looked similar to TI's so I tried that, but didn't work out either.  The inductance shot up to 29.9uH.  I should also note that I am measuring the inductance with an L/C meter but it doesn't put out 1Vrms as it is called out in the datasheet.  I am not sure if this is the root cause.  The L/C meter I am using is UCTRONICS L/C meter (https://www.amazon.com/UCTRONICS-Precision-Capacitor-Inductive-Capacitance/dp/B0856RCWBJ/ref=sr_1_1?dchild=1&keywords=UCTRONICS+High+Precision+Handheld+LC+Inductor+Capacitor+Tester%2C+Inductive+Capacitance+Meter+Kit%2C+1pF-100mF%2C+1uH-100H&qid=1623101912&sr=8-1).  It does a decent job measuring accurately the Inductors I have presented to it.  I have not measured the output voltage it puts out.  Could this be the issue?

  • 0
    TI__Mastermind 23410 points

    Hi Roberto,

    Would you be able to show the waveform at VRECT? 

    Also, feel free to take a look at the FAQ for wireless power receivers: https://e2e.ti.com/support/power-management-group/power-management/f/power-management-forum/910420/faq-startup-troubleshooting-for-wireless-power-rx

    Best Regards,

    Anthony Pham

  • 0 in reply to Nick Brylski
    Intellectual 800 points

    Thank you... I have seen those posts but unfortunately, I have not gotten mine to work.

  • 0 in reply to Anthony P Pham
    Intellectual 800 points

    Anthony,

    I have put the VRECT waveform reaching 10V.  Can you describe the proper way to measure Ls'?  do you need to put my device on top of the TX charging pad and measure?  I don't have a 1Vrms L/C meter. I am using this one https://www.amazon.com/UCTRONICS-Precision-Capacitor-Inductive-Capacitance/dp/B0856RCWBJ/ref=sr_1_1?dchild=1&keywords=UCTRONICS+High+Precision+Handheld+LC+Inductor+Capacitor+Tester%2C+Inductive+Capacitance+Meter+Kit%2C+1pF-100mF%2C+1uH-100H&qid=1623101912&sr=8-1

    Does a decent job.  Anyway I PLA enclosure, coil faces down, ferrite faces the battery, above the LIPO battery is the PCB.  Ls' was measured this way and I have posted the calculations.  Then I read a different vendor's IC and they have identical equations, but they say to measure Ls' to place my stack on the TX pad - I didn't turn on the TX pad.  While the TX pad was off, I measured Ls' and measured slightly higher .  Using that measurement re-adjusted C1 to around 84nF, and Ls' = 29.9uH in that condition.  Still the same issue.  it will not lock.  The TX Pad just blinks and blinks.  Below is a capture of VRECT (magenta), yellow is OUT.  it won't come up.

  • 0 in reply to Roberto Rodriguez
    TI__Mastermind 23410 points

    Hi Roberto,

    Thanks for the update and I see you've updated the thread with more information. 

    The VRECT swing does look high but as it does not exceed the 15V OVP point, this should be fine. This could be a communication error. 

    We can try and adjust the COMM 1&2 capacitors. Can you try the following values:

    • Changing COMM 1&2 capacitors to 47-nF?
    • Changing COMM 1&2 to 10%, 20%, and 30% of the series capacitor (C1)?

    Best Regards,

    Anthony Pham

  • 0 in reply to Anthony P Pham
    Prodigy 90 points

    Hello Anthony... I replaced the capacitors COMM1 and COMM2 to 47nF  Here are the results.  About 14V in VRECT. The pad continues blinking and won't lock solid LED. I will make the following changes and try

    I will post the results with

    10%, 20% and 30% of C1 which at the moment it is set to 84.2nF (82nF||2.2nF). 

    10%*C1 = 8.42nF

    20%*C1 = 16.84nF

    30%*C1 = 25.26 (closest 22nF||3.3nF 25.3nF)

  • 0 in reply to Anthony P Pham
    Prodigy 90 points

    Hi Anthony,

    Here is VRECT (Magenta), VOUT(YELLOW) with COMM1, COMM2 @ 10% of C1 or 8.2nF in both sides.  It does not lock

  • 0 in reply to Anthony P Pham
    Prodigy 90 points

    Hello Anthony...

    Here are the VRECT for 20% of C1 or 16.4nF on COMM1 and COMM2.  The charging pad still blinks and sometimes I noticed the VRECT pulse time was longer than 81msec than what is shown in the image- in this condition the charging pad was blinking faster short pulses see below. Also to get it in that mode, I had to lift my device from the pad Up and Down several times waiting in the air for about 5seconds, the placing it on the pad.  It only happens intermittently and it is not repeatable.  It was hard to capture the second picture.

    QUESTION

    Right now there is a load resistive on the output of the BQ510003 OUT pin to ground pulling 5V/25Ohms = 0.2A.  If there was no load, does the charger still turn ON - Assume all is perfect.  I am wondering what happens when the load current reduces in the device to around 10mA or less.

  • 0 in reply to Anthony P Pham
    Prodigy 90 points

    Hi Anthony,

    Here are the 30% *C1 or 25.3nF on COMM1 and COMM2.  The OUT comes up to 5.4V.  tried raising the device off the pad to see if I can get it to lock.  Maybe the H field is too strong now and it is shutting off.  Not sure.

    In this case the waveform looked different, and OUT came up, where in the other cases, it didn't.

  • 0 in reply to Anthony P Pham
    Prodigy 90 points

    Hi Anthony,

    Is there any other test I can perform?  Have you looked at the VRECT measurements?  The unit doesn't lock on the charging pad despite of having done all the C1,C2 component updates, COMM1, COMM2 updates.  I have placed a 10K to ground on TS-CTRL, by passed the thermistor just to make sure the device is shutting down due to any temperature. 20K resistors from VRECT to FOD and set the RLIM to 200mA approximately.  The coil used is 27uH, measured Ls on free space (outside coil by itself), measure Ls' two way.  One way measured Ls' inside the enclosure, and the other way I measured it, Coil in the enclosure but placed on the charging pad.  I asked which is the correct way, and I have not gotten clarification yet.  The diagram in the datasheet is not clear to me yet.  I don't know if I have to build a fixture with those specs.  Other vendor with similar ICs say to place the device on the charging PAD to measure Ls'. 

    Please advise on what else I can try.

  • 0 in reply to Robert Rodriguez
    TI__Mastermind 23410 points

    Hi Robert,

    Robert Rodriguez said:
    One way measured Ls' inside the enclosure, and the other way I measured it, Coil in the enclosure but placed on the charging pad.  I asked which is the correct way, and I have not gotten clarification yet.

    It'd be best to measure with the housing/enclosure. It looks like the device is best working in that second image where you say it was difficult to measure. What is the meaning of the USB flash drive detected? Is this an oscilloscope feature or what occurs in your application?

    Would you also be able to show a wider scope shot of this? It looks like the RX is still communicating and it takes a few ms before the output fully comes on. 

    The Ls' measurement should be measured on top of a primary shielding not a charger pad. 

    Can you make these measurements and provide the new calculation for C1?

    Robert Rodriguez said:
    Right now there is a load resistive on the output of the BQ510003 OUT pin to ground pulling 5V/25Ohms = 0.2A.  If there was no load, does the charger still turn ON - Assume all is perfect.  I am wondering what happens when the load current reduces in the device to around 10mA or less.

    Yes, you can find Figures 35 to show the operation with no load.

  • 0 in reply to Anthony P Pham
    Prodigy 90 points

    Hi Anthony,

    I will go back to the COMM1, and COMM2 at 20% of C1.  I will measure Ls' in the stack shown below for my case not on top of the charging PAD.  I have done that way too, but I will do it again.  The calculations are posted in the first message for reference.

    Below is a cross section of the stack.  I also have ferrite material sheet I can add between the coil and battery for a second test to see if that would help.

  • 0 in reply to Robert Rodriguez
    TI__Mastermind 23410 points

    Hi Robert,

    I appreciate the info and look forward to your results!

    Best Regards,

    Anthony Pham

  • +1 in reply to Anthony P Pham
    Prodigy 90 points

    Hi Anthony,

    I tried again, added between the coil and the battery 38M3010AA1212 (1934-1086-ND Digikey) ferrite sheet the size of the battery. Peeled off the adhesive, and attached to the battery surface.

    Measured the Ls' once more and the calculations are attached below.

    C1 worked out to be 94nF, and C2 940pF with Ferrite in between coil and battery.  The measurements are shown below.

    I didn't change COMM1, COMM2 left it the same as 22nF.  I only updated the C1, C2 components and this time I see the CHG status LED ON and provides 5V output as expected.

    This was very difficult to figure out.  I had a hunch that more ferrite was needed to isolate the battery and board from the coil given they are in close proximity of one another.   Hopefully others can benefit from this extensive troubleshooting, ultimately resulting in success.

    Regards,

    Robert