• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 78 subscribers
  • Views 288 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

CCS/MSP430FR2512: CCS error: MSP430: Error connecting to the target: Unknown device when programming the chip a second time

Gian Moreira
Prodigy 20 points
Part Number: MSP430FR2512
Other Parts Discussed in Thread: MSP430FR2311, ENERGIA

Tool/software: Code Composer Studio

Please, help me figure out why the CCS is giving me the following error:

MSP430: Error connecting to the target: Unknown device

I am currently trying to program a custom made PCB using the MSP430FR2311 Lauchpad as a programmer for the MSP430FR2512. I am simply connecting the SBWTCK to pin 3 and SBWTDIO to pin 4 of the FR2512. Whenever I program the PCB, I use the 3V3 available from the LP. Both devices also share the same ground pin.

I am using a 1nF capacitor connecting the RST pin to ground, and a 100k resistor pulling the same pin up.

The strange part is that I first ran a test on the MCU on the PCB to unsure that all pins are working, so I simply set all pins as output and made them toggle at 8Hz. This first program loaded without a problem, and whwnver I hook up power to my PCB, I can see that all pins are still toggling at 8Hz. However, when I am trying to load the actual firmware that should run on that device, I get the following error from CCS. I did not change anything on the pins.

I wonder if the fuse could be blown, so I tested with a multimeter. I measure an 8Mohm resistance to ground, and whenever I connect the test pin to 3V3, I can measure 70uA going into the pin.

I don't know if the problem is if the fuse, specially considering that the MCU was not in contact with more than 3V.

Do you know what could be the error?

  • 0
    TI__Genius 17960 points

    Hello Gian,

    the fuse in MSP430FR2xxx is not a real physical fuse, but a register settings and configuration, locking the JTAG/Spy-Bi-Wire interface. So you cannot check it by measurement. We are calling it for legacy reasons still fuse, since in some older families of MSP430 it has been a real physical fuse.

    Theoretically the fuse could be set, if you have configured this in CCS, but usually this cannot happen by coincidence, as the default setting is not locking the fuse and the sequence of actions is pretty complex to get there.

    Looking at the information you provided, my suspect number one would be the reset circuitry. you are using a pretty high pull-up value of double what we typically use and still 1nF capacitor. This will degrade the timings. For debugging I would anyway recommend removing a reset capacitor completely.

    The second suspect would be the length of the debug connection and its signal integrity. I would definitely try to keep this as short as possible.

    Best regards

    Peter

  • 0 in reply to Peter Spevak
    Prodigy 20 points

    Hello Peter,

    I tried running with a lower resistance with no success. When I removed the capacitor, the waveform coming from the SBWDIO looked triangular on the oscilloscope.

    However, I got the same problem with the FR2311, but I was able to reprogram it after with Energia.

    After running the Energia code, I was able to switch back to CCS. Since Energia doesn't support the FR2512 series, I can't test on it.

    I am wondering what Energia did during the programming that cleared the MSP430FR2311.

    Do you know if there is any difference during the programming stage?

    Regards,

    Gian Moreira

  • 0 in reply to Gian Moreira
    TI__Genius 17960 points

    Hello Gian,

    many thanks for the update.

    The signal on the SBWDIO definitely should not look like triangular, at least not during the phases, when a driver is active. I would recommend checking the HW setup. Something seems to be wrong there.

    In terms of the effect of Energia, this is also sort of surprising to me, as the low level drivers used for the debug communication should be the same, and I assume you're using the same HW setup. What of course can be the case is that some timings differ from what the C-Spy debugger of CCS is applying in terms of signals might differ. This also sort of points to some issues with marginal HW setup.

    Best regards

    Peter

**Attention** This is a public forum