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.

Booting control stick from flash

Hi,

I am having trouble getting the 28027 on a controlStick to run standalone. The problem is that whenever the processor is powered so is the FT2232 and this holds the TRSTn input high. The boot loader sees this and sits in Emulation boot mode and will not branch to the flash start address (0x3F7FF6). Is there any way of solving this other than lifting the output pin on the FT chip?

Cheers,

John

  • As long as you do not start CCS and connect to the target, the line /TRST from the FTDI chip will stay low - the default boot mode is Flash. You can still use the power supply via USB and the the Stick will start from Flash memory. I have attached an example for a blinking LED from Flash.  

    28027_Lab1_3.pdf
  • Hi Frank,

    Thanks for your reply. However, my findings do not support what you say: I have scoped up the TRSTn line (at R10 on the control stick) along with the 3v3 supply. The TRSTn goes and stays high about 5ms after the 3v3 supply. I am powering the board from a powered USB hub that is not connected to a PC, so CCS cannot be connected.

    If I connect the stick to a PC, run CCS, connect to the target, reset the CPU, poke 0x3F7FF6 into the PC register (jump to Flash branch instruction) and then run, the code runs OK. This means that my code is correctly programmed into flash and runs OK. The problem must be in the booting logic.

    Am I doing something else wrong? Or do I have a faulty control stick?

    Cheers,

    John.

  • Hi John,

    I am not sure what happens with your setup. We use the sticks in a small series of measurement devices and boot directly to Flash - no PC and no CCS are present. The sticks are powered either by a USB cable or even by an external 5.0V power supply (stick PIN16).  I haven't scoped the start sequence of TRST, but I'd guess that even with your 5ms high-signal the stick should start its operation in standalone mode after TRST goes to 0.

    Regards

     

     

     

     

     

  • Hi Frank,

    Thanks for your further thoughts. Yes I would have thought that 5ms is enough time for the boot code to get past the bit that checks the level of TRSTn, so I looked further: It turns out that the GPIO37 signal (which is also JTAG TDO) also remains low for the same 5ms before going high with the TRSTn signal. This means that my processor is probably attempting to do a SCI boot explaining why my code in flash does not run. I also notice that my board is fitted with 2k2 for R9, while the circuit diagram I have says it should be 3k3 - I wonder if the 2k2 is too strong a pull down? What value resistor is fitted on your boards?

    It is hard to debug because I cannot determine what the processor is doing without connecting the emulator - and doing that  that changes the boot conditions. Perhaps this should be known as the Piccolo uncertainty principle?

    Anyway, I am going to have a go at removing R9 completely to see if that gives any joy.

    Thanks again for your help.

    Cheers,

    John.

  • Yes this worked.

    I removed R9 from the control stick and it will now boot from flash correctly when powered from my USB hub. Further, I can still connect to the processor in CCS with no problems.

    All in all, I am not sure what this resistor was there for (perhaps to force an SCI boot from the other channel of the FTDI chip?), but removing it has solved my problem.

    Cheers,

    John.