Wireless transmitter not starting

I may have a similar problem.

I've implemented a design similar to that used on the BQ500210, except that I do not have the Low Power Supervisor MSP430G2001.   I've left the pins floating that would have been connected to the MSP430G2001.

I see the BQ500210 is pinging the TX coil with every 15 seconds putting out a burst for approx. 75ms.   This is the case when the RX coil is on top of the TX coil and when the TX coil is just sitting there.

I do not have the pad on the back of the BQ500210 grounded - this was a layout error and I'm in progress with a board spin to fix that.  I've read the recent app note with recommended design practices and for the most part I've followed the rules, except I've got a 2 layer board with reasonable ground plane.

I do not want to spin the new board without knowing the root cause of the failure of the WP TX.    The fall back will be to replicate as exactly as possible the design and layout of the EVM.  

Are there recommendations for

1) not using the MSP430G2001 and what to do with the pins?

2) debug techniques

I'll add more details on what's been tried and circuit measurements if there is a response.

Thanks in advance,

Thom

A number of things can cause this type of problem on a new layout.  See SLUA635A for information on layout and debug.   The risk of problems with 2 layer boards are high.

Thanks for the pointer to SLUA635A.   I had done this layout before the app note was published and had (luckily) followed good practice and for the most part obeyed the recommendations except on the 2 layer decision.

In making the updated 4 layer PCB I've also gone thru my old layout carefully and discovered the following problems:

1) I did not ground on the other side of the 10 ohm resistor the COMM- signal

2) I had the foot print wrong on the BAT54SW so it was killing the COMM signal

3) I had the input polarity reversed on the INA214

4) I do not have the power pad grounded on the back side of the BQ50210 due to hand assembly (have not corrected this one yet).

I have been able to verify the power train on my 2 layer board by hacking in the PWM signal from the BQ50210 EVM and have been able to transfer power.

So in short, I had errors in my layout and schematic affecting the communication path and current sensing which explains why the BQ50210 is pinging but not starting up.

With care and attention to details in checkplot, following the guidelines in the layout, and having a small run machine assembled I'm hoping my power transmit startup problems will be gone.

Thom

One more question - Do I need to connect PMB_CLK and PMB_DATA thru a 10K resistor to 3.3v as shown on the EVM or can I leave them floating (un-connected).

Thanks,

Thom