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.

BQ500212A can't work with BQ51013B

Other Parts Discussed in Thread: BQ500212A

I am testing a BQ500212A  board designed myself these days. I've got some problems.

The main issue is that even though the Rx is placed on the Tx coil, it still doesn't start charging. It seems like the Hand-Shaking process has failed. The BQ500212A switches the TLV70033 on and off contently, just like there is no Rx on the coil surface. 

I also have some queries 

1. Is it possible that the pulse time is too short for the communication between Rx and Tx? I measured the waveform of the RC circuit and found that the pulse time is about 85ms and the interval between two neighboring pulse is about 450ms

2. Will it influence a lot if I use COMM resistors with accuracy of 5%? The waveform between COMM+ and COMM- shows that the communicating signals is weak, I am not sure is that normal. 

3. I imitate the PCB layout from the 500212A EVM board.

My brain is very chaotic. That's all descriptions I can present. I really want and appreciate someone can give me some help. If you want to see the waveform, please inform me where I should measure. 

Here are the schematic and gerbe files:

6813.OUTPUT.rar

Thank you for your patient! 

  • Hi, we cannot open .rar file, can you attach the schematic as .jpg? Also can you capture the scope waveform of
    - RECT and VOUT
    - AC1/AC2 signal

    Also, do you know if the RX is proved good?

    Thanks,
    Wenjia
  • Thank you for your reply! 

    I think the Rx is good because it works well with a Tx designed by others. 

    Here is my Tx schematic: 

    Here is waveform: 

    1. RECT:

    2 VOUT

    3 AC2

    4 COMM

    That's all.

    Thanks!

  • Hello,

    This looks like a packet decoding issue of the TX. Bill from the team will reply with his comments on it.

    Thanks
    Wenjia
  • Hello 

    This looks like a communication problem on the TX side, cannot demodulate the information the RX is sending.  In the scope captures you can see the RX is sending signal strength packet and larger part of the ID packet, but TX stops power transfer. 

    When the TX decodes a communication packet it will keep power on the RX to continue decoding additional packets.  In poor communication you will see a digital ping that will vary in width as the TX has varying success decoding the info.  But what we see here is consistently the standard digital ping indicating the communication is very bad or broken.

    See below review and comments.

    My most likely suspects are the two layer PCB and Reversal of PWM_A and PWM_B.

    1.)   Is it possible that the pulse time is too short for the communication between Rx and Tx? I measured the waveform of the RC circuit and found that thepulse time is about 85ms and the interval between two neighboring pulse is about 450ms

    • No—The pulse time is satisfactory. 

    2.)   Will it influence a lot if I use COMM resistors with accuracy of 5%? The waveform between COMM+ and COMM- shows that the communicating signals is weak, I am not sure is that normal. 

    • 1% is recommended but is not the cause of this problem.

    3.)    I imitate the PCB layout from the 500212A EVM board.

    • The EVM PCB layout is a 4 layer board this is required for low noise and reliable demodulation of communications packets.
    • In reviewing your layout it looks like a 2 layer board, can you confirm this?
    • We have seen a number of two layer boards and their performance has been very poor.

    Also:

    Communications packets demodulation requires sampling point to be synchronized with switching signal.  PWM_A signal should drive the coil side ( L1 ) of the H-Bridge and PWM_B signal should driver the capacitor bank side ( C26, C27, C28, C29 ).  This appears to be reversed on you schematic.

     

    Some other areas to look at:

    1.)   Grounding of bq500212A

    • Vias under the IC connecting thermal pad go large ground layer
    • Thermal Pad used as a common tie point for all IC grounds
    • Solid internal ground plan connecting IC to 3.3V converter and COMM sample ckt (R36, R40, R35)

    2.)   Vcc filtering of bq500212A

    • V33D capacitors located close to Pin 33 with very good ground return to pin 33
    • V33A capacitors located close to Pin 34 with very good ground return to pin 36
    • Ground return for Pin 33 and 36 should be top layer with no Vias

    3.)   Routing of Comm + or Comm-

    • Try to rout the signals close to each other
    • Comm – (R41) should be connected to ground at R40

    4.)   Noise on Comm ckt

    • Check grounds of R40 and C25 should be very good grounds
    • We have found that Noise on 3.3V at R35 will degrade Comm, try a high freq filter capacitor from R35 to gnd.
  • Thank you so much! I have already solved this problem!
  • Thank you so much! I have already solved this problem! It seems that the Reversal of PWM_A and PWM_B caused this problem. The device worked well after I exchanged the PWM_A and PWM_B.