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.

CC3220SF: Migrating from CC3220SF Launchpad XL Development Board to CC3220SF IC

Part Number: CC3220SF
Other Parts Discussed in Thread: UNIFLASH, TMDSEMU110-U

Hello,

I am working on a senior design project in university right now, and we are using the CC3220SF LaunchpadXL development board currently.  The project involves WiFi communication and analog GPIO inputs fed into the internal ADC.  We have our project working with the development board and now we want to migrate over to a custom PCB in which we will only use the CC3220SF MCU and the minimal necessary hardware for our project.  I have been reading the datasheet (CC3220R, CC3220S, and CC3220SF SimpleLink™ Wi-Fi® Single-Chip Wireless MCU Solutions datasheet (Rev. B) and have been looking at page 70 which shows a typical application of the CC3220SF on a custom PCB.  I found a few things confusing and was hoping someone could clear them up.

  1. Is there an internal oscillator in the CC3220SF MCU?  The product page says "40.0-MHz Crystal With Internal Oscillator" and also says that the ARM Core is clocked at 80MHz, and page 70 of the datasheet shows an external 40MHz crystal oscillator (Y2), which I found confusing.  If there is an internal oscillator, would we still need an external one on our PCB?
  2. Is there internal flash in the MCU?  The datasheet refers to an "Optional 1MB of Executable Flash" and "External Serial Flash" multiple times, and both are classified under "embedded memory" although page 70 of the datasheet shows external flash in the diagram (U2).  The LaunchPad XL development board has external flash on the board.  When I use Uniflash to program the flash, does it program internal or external flash?
  3. Connection to a computer: The LaunchpadXL development board has a "Back-channel universal asynchronous receiver/transmitter (UART) through USB to PC" which is how it connects to my laptop.  How can I create this connection using a custom PCB?  The LaunchpadXL development board uses an onboard XDS110 JTAG emulator which handles programming and debugging and communication back to the PC.  How can I implement this on a custom PCB so that software such as Code Composer Studio and Uniflash still work?
  4. For the project I am working on, which uses WiFi and GPIO pins as analog inputs, are all of the pieces of external hardware on page 70 of the datasheet necessary for the system to work?
  5. Any other tips for migrating from the development board to the custom PCB?

Thank you for your help.  Since I am a beginner, I am still learning much of this and any direction would be of great value to me.

  • Hi,

    Answer to you question:

    1. CC3220SF QFN chips requires two external crystals (32kHz for RTC, 40MHz for generating main clock). Both crystals are mandatory. By internal oscillator is meant part of chip which generates clock from external XTAL.

    2. Yes, chips with SF (e.g. CC3220SF) have internal 1MB flash used as a execution cache. External SPI flash chip (sFlash) is still mandatory for all CC32xx chips. By Uniflash is programmed content of external SPI flash.

    3. For developed purpose you can connect your board to XDS-110 from LaunchPad (JTAG + UART). Other option can use external debug probe TMDSEMU110-U.

    4. Yes.

    5. Maybe you should consider usage of module (e.g. CC3220MODSF or CC3220MODASF). CC32xx QFN chips are very sensitive for a proper layout design. Design with modules is much easier and you have will have better chance to success at first revision of your board. After you will have done design of your board, you can ask TI for design review here.

    Jan

  • Hi Jan,

    Thank you for your response.  That was very helpful and I think the CC3220MODASF would be great for my project. 

    I am trying to come up with an intermediate developing and debugging environment, in between using the LaunchPad XL dev board and using a custom PCB.  I would like to test the CC3220MODASF on a breadboard so I can experiment with the peripheral connections and ensure that the circuits are working before designing a PCB.  Before, when I was thinking of using the CC3220SF chip, I was considering using a product like this (204-0018-01 Schmartboard, Inc. | Prototyping, Fabrication Products | DigiKey) to breakout the pins of the VQFN chip.  However, the CC3220MODASF is a unique shape and I have not found any QFN to DIP adapters for it.  Any advice on how to break out the pins or how to test on a breadboard?

    Additionally, when it comes to soldering the CC3220MODASF, I see that there are 9 pads in the middle on the bottom of the chip, whereas QFN chips usually have one pad there.  I have read a lot about the difficulties of soldering regular QFN chips because of this pad and I am concerned about soldering the CC3220MODASF with 9.  I also see that the typical application of the CC322MODASF (page 71 of the CC3220MODASF datasheet) shows that all 9 of these pads should be connected to ground.  When designing the PCB, could I connect all of these pins together and then connect them to ground?  Alternatively, would it be okay if there were a solder bridge across these 9 pads?

    Thank you for your help.

  • Hi,

    I am not aware about any commercially available adapter board. But maybe you can check this thread.

    I think best example of proper layout are design files for CC3220MOD LaunchPad. I think reason why thermal ground is split into 9 pads is a better re-flow under module. Under module is solid ground plate with gaps in solder mask.

    Jan