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.

TPS65981: Blank SPI external flash memory programming

Part Number: TPS65981
Other Parts Discussed in Thread: , TPS65987D, TS5A3159

Hi,

May I use the DP-EXPANSION_EVM board as a programmer for the serial Flash Memory?

I intend to build a board containing a socket for the serial flash memory, connect it to the DP-EXPANSION_EVM and program the flash using the Application Customization tool?

Is it feasible?

Another question: 

In order to ease mass production, I wonder if my board coud contain all the h/w needed to program the serial blank flash memory.

If the serial flash could receive SPI from TPS65981 and also the SPI from a microcontroller (each of them 3-stated in suitable time), the microcontroller would program the blank flash and then release the control of the serial flash for the SPI interface of TPS65981.

Does .SPI_MOSI and SPI_CLK go to 3-state?

Best wishes,

  • Hello Newton,

    Unfortunately, you cannot use the Expansion EVM as a programmer for the serial flash. It was not designed with that functionality in mind.

    In theory, this shouldn't be a problem. You can reference the TPS65981EVM, and how the FTDI is connected to the SPI flash as this will be a similar design to what you are trying to accomplish.

    Let me know if these answer your questions or not! If so, please click, This resolved my issue

  • Hello Adam,

    Thank you for the answer.

    As for SPI_MOSI and SPI_CLK pins of TPS65981 going  to 3-state, do you have any information? This would ease the design of this "shared" SPI.

    Best wishes,

  • Hello Newton,

    Would you mind elaborating on "3-state" and how it would help ease the design? Have not heard this term before, so I wanted to get a clear definition from you before I state whether the device can use it or not.
  • Hello Adam,

    A Tri-state Buffer can be thought of as an input controlled switch with an output that can be electronically turned “ON” or “OFF” by means of an external “Control” or “Enable” ( EN ) signal input. This control signal can be either a logic “0” or a logic “1” type signal resulting in the Tri-state Buffer being in one state allowing its output to operate normally producing the required output or in another state were its output is blocked or disconnected, in other words, high impedance.

    So if I added an extra SPI interface connection between my controller and the external flash memory (keeping the SPI connection between TPS65981 and the external memory flash), the extra SPI would be used only to program the blank memory, and it would be kept in three-state mode (high impedance) after that initial programming procedure, letting TPS65981 SPI interface take the control. from that moment on.

    But TPS65981 SPI couldn't interfere during the extra SPI programming phase, that's why I asked about the 3-state feature. If TPS65981 SPI is not 3-state featured, Id have to add a 3-state buffer between TPS65981 SPI and the external flash memory, to isolate them during the first programming phase.

    Best wishes,

  • Hello Newton,

    Thank you for sharing that. Usually I am the one to teach people new information here on the forms but you taught me something which is a welcomed change.

    Unfortunately, the TPS65981 does not have an internal tri-state buffer to control SPI programming. However, wee are currently working on a TI-Design that will have two TPS65987D programed by a single FTDI. The SPI lines from the FTDI lines are connected to both '87D, excluding the CS line. The CS line from the FTDI goes to an external header pin, and depending on what device the user is wanting to program, they connect this header pin to another using a connector.

    You could do a similar approach using an SPDT switch such as the TS5A3159. You could have one throw connected to the CS for the microcontroller, and the other throw connected to the CS of the TPS65981. The pole could then be connected to the CS line of the FTDI, and the switch port can be controlled by a switch port from the microcontroller.

    Let me know what you think of this idea as opposed to the tri-state buffer