CC8530: power button and paring button not working on cc8530 slave

Hi Austin,

You can submit your board design to SIMPLELINK-2-4GHZ-DESIGN-REVIEWS for further review, otherwise you can read more about CSN/GIOx operation from SWRU250.

Regards,
Ryan

Ryan,

I have submitted the design files. I also had to rewrite the device definition file for TAS2562 to allow two amps to be used. The cc8530 user guide says one chip can control multiple audio devices, but the purepath wireless configuration does not have a way to do so built in. I forgot to include the device definition file originally, so I can also email that if necessary.

Thank you,

Austin

I do not expect the device definition file would have affected behavior of the CC8530 power and pairing buttons.  It is necessary to isolate whether this is a hardware or software issue.  Are you able to run default projects on the CC8530 without issues (note that this will not be using your TAS2562 amplifier configuration)?  Or can the CC8530 be operated as expected after physically isolating it from the on-board TAS2562s?  Also note from the CC85XX User's Guide that "If no audio network is found or a protocol slave loses connection to the protocol master, it can automatically reduce audio device power consumption and/or power down after configurable timeouts".  Thus you might consider making sure that the boards are able to connect.  You can also review the Status LED Indication states to make sure that "ALONE" is not too similar to "PAIRING" such that it could go unnoticed for your system.

Regards,
Ryan

I ran default projects and it did not fix the problem. I don't have a good way to physically isolate it from the tas2562 except from cutting the trace, which I would prefer to hold out on doing if possible. I changed the button debouncing configuration a lot, but that did not seem to help either. I eventually found a stupid little mistake in the hardware: I had two digital voltage supplies (one 1.8V and one 3.3V). The CC8530 was connected to 3.3V digital supply, but the I2c bus and tas2562 were on a 1.8V supply. Is it possible having the i2c bus too low would cause the chip to not start up correctly even with default projected? I can try disconnecting the 3.3V supply and soldering wires across the board to make both digital supplies 1.8V, but I have not had the time to test this yet.

Thank you for the update.  This debug effort appears to confirm a custom hardware issue, I'm not sure whether or not it concerns the I2C bus but all avenues should be considered.  How were the board components assembled/soldered, i.e. in-house or by a fabrication house (possibly the same company who created the PCB)?  Do you have an oscilloscope with which you could monitor the C8530 pins?  I recommend having another engineer review your physical setup and connections.

Regards,
Ryan

The PCBs were assembled by PCBway, but there were other mistakes that required me to rework a couple components. I had ordered 3 boards, two of them are now fried, one I delaminated a pad while trying to rework one, the other is what I had been testing.

I originally had pulldown resistors for active high buttons in the design, since I missed in the user guide that the power button on CSN must be active low. I had been testing by connecting CSN to VDD then briefly disconnecting to allow the pulldown resistor on the board to act as the button push. This worked when I was testing the transmitter board, but I noticed on the receiver that with the pulldown resistor it was leaving the pin at 0.6V instead of 0V because of a higher leakage current than expected. I couldn't find the maximum value for the pin to register as low anywhere in the data sheet or user guide, but think this could be the problem. This morning when I tried removing the pulldown resistors I somehow messed up the 3.3V LDO and it started outputting the full 5V supply voltage, which I can only assume fried the CC8530 since it is so far above it's max supply rating.

So I can either send in the last prototype board to be looked at or try reworking and testing it myself. I do have a scope, but am not confident I could rework the last board successfully, as it requires replacing an LDO with an x2son package. I think I got lucky it worked for a while the first time.

Unless you think sending the board in to be looked at is necessary to find the problem, I'm ready to correct the errors on the board I am already aware of, have the design reviewed, and try again. I think the buttons are likely not working because of one of the mistakes I am already aware of, but do not know for sure.

Thank you,

Austin